Estudo da extração de As(III) e As(V) em sistemas aquosos bifásicos: uma nova metodologia para determinação e especiação de arsênio inorgânico

Abstract

In the present work, a new method for the extraction, speciation and determination of inorganic arsenic was developed, using aqueous two-phase system (ATPS) technique with the extractant ammonium pyrrolidine dithiocarbamate (APDC) and analysis through Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES). Inorganic arsenic species, As(III) and As(V), are considered toxic and carcinogenic, especially in contact or ingested at levels higher than those permitted by legislation. A simple and efficient approach for the extraction of several analytes without the use of organic solvents is the Aqueous Two-Phase System (ATPS) technique. In this study the following experimental conditions were optimized: pH of the reaction medium; nature of ATPS-forming electrolyte and the polymer; type and concentration of extractant; the tie-line length (TLL) and the mass ratio between the ATPS top and bottom phases. The highest percentage of extraction value (%E = 98%) was obtained for the ATPS composed of L64 + Na2SO4 + H2O at pH = 6.00, in TLL equal to 33.55 % w/w and using APDC in a molar ratio (APDC/As(III)) of 960. Under these conditions As(V) was not extracted satisfactorily (%E = 18%) to ATPS top phase, showing the great potential of the system for the speciation of inorganic arsenic. Moreover, the method was validated according to Brazilian Sanitary Surveillance Agency validation guide (Resolution nº 899, from May 29, 2003). The following analytical parameters were determined: linear range, limit of detection (LOD), limit of quantification (LOQ), accuracy and precision. Linear range between 0.66-80 g kg-1 of As(III) with a coefficient of determination of 0.9957 was obtained. The method showed LOD and LOQ of 0.20 and 0.66 g kg-1, respectively, which are according to the limits required by legislation. The relative standard deviation (RSD) varied from 4.88% to 9.45%, and the RSD in different days was 11.1%, which were also according to the limits of the validation guide. In the accuracy studies, the method presented relative error rates ranging from -7.44% to 9.90% and recovery rates ranging from 92.6% to 110%.