Avaliação de carvões ativados para extração em fase sólida : um método alternativo para determinação de desreguladores endócrinos

Abstract

This work aimed the development of an alternative method of extraction/preconcentration associated with an affordable analytical technique for the detection and quantification of the 17a-ethinylestradiol (EE2), 4-n­ noniphenol (NP) and bisphenol A (BPA) analytes, usually classified as endocrine disrupters . Activated carbons were employed as adsorbent material and the chosen analytical technique was the high performance liquid chromatography coupled with ultraviolet visible spectroscopy (HPLC-UV-VIS). In this context, three types of activated carbon were evaluated as possible adsorbents for the construction of a solid phase flow extraction system. The carbons were assessed through the comparison between their surface areas (m2 g-1), their pore volumes (em- g-1) and their adsorption capacities for 17a-ethinylestradiol. The carbon PA 6:1 carbon presented the largest surface area and adsorption capacity (500 mg g-\arbon) among ali carbons tested, and was chosen to be the adsorbent material used on the development of the present extraction/preconcentration method. The flow extraction/preconcentration method was optimized an experimental design, with the following variables being investigated: preconcentration and elution flows, the amount of adsorbent material and the composition of the eluent. For each of the investigated analytes, the validation of the method succeeded its optimization for the determination of the following parameters: linearity, preconcentration factor, detection limit and quantification limit of the technique and of the analytical method, precision and accuracy. Ali of the three preconcentrated calibration curves met the ordinary least squares (OLS) method premises for the proposed working range (O to 1 000 ~g L-1). The calculated preconcentration factors (3.16 for EE2; 3.53 for NP and 2.77 for BPA) are low when compared to those achieved by solid phase extraction with commercial cartridges, possibly due to the strong interaction between the analytes and the activated carbon employed as adsorbent. The determined quantification limits of the method ranged from 22 to 38 Ug L-1, which is considered high, making it inadequate for environmental monitoring. Nevertheless, the described method may be employed, for instance, in the evaluation of the efficiency of pollutant treatment/removal processes, where higher concentration levels are often utilized. The accuracy was assessed according to the European Comission acceptance criteria (recoveries between 80 and 110 %), thus, only the recoveries achieved for the first two investigated levels of 17a-ethinylestradiol (130 and 120 %) did not present proper agreement. The precision was expressed through relative standard deviations, which ranged from 2.7 to 9.6 % and are also in agreement with the acceptance criteria defined by the European Commission, which establishes a maximum relative standard deviation of 10 % for concentrations over 100 Ug L-1.