Effect of foreign substances on the determination of phenols total content using single-wavelength and multi-wavelength UV-spectrometry

The selectivity of spectrometric determination of toxic phenols total content (cΣ) was tested. Water solutions, simultaneously containing up to 5 different phenols, were analyzed by two methods, based on phenols conversion to certain azodyes under the action of diazotized sulfanilic acid (DSA). Method A included the measuring of generalized signal of phenols at 380 nm and estimation of cΣ in terms of phenol (C6H5OH). Method B included the measuring of generalized signals at 10 wavelengths in the range 360-410 nm and calculating of cΣ with inverted multivariate calibration. In the absence of foreign substances (Z) method B gives much more accurated results than method A. To study the effect of certain Z, different quantities of Z were added to the phenol mixture, cΣ-values were determined using both methods, and the arising addition systematic errors were compared. Introduction of a 10-fold excess of arenes, acids, bases, salts of alkali and alkaline earth metals does not affect the generalized signals and the results of cΣ determination. Primary arylamines react with DSA similarly to phenols; it leads to exceedingly overestimated results of сΣ determination (even at cz/cΣ = 1), regardless of the method choice. Humates showed the similar effect. On the contrary, the presence of strong oxidants and especially the presence of active reductants lead to underestimation of cΣ, probably due to destruction of DSA. The excesses of Cu2+ and Al3+ lead to underestimation of cΣ too, obviously due to sorption of azodyes with forming hydroxides. As a rule, the influence of Z was more pronounced when method B was used for phenols determination, with all other things being equal. Possible methods for preventing or taking into account the corresponding systematic errors in assessing the phenolic contamination of natural and waste waters are discussed.