Contribution of system components to dispersion in the analysis of micro-volume samples by flow injection flame atomic absorption spectrometry

Abstract

Response characteristics of the atomic absorption spectrometric detector and dispersion contributions of individual components of the flow injection atomic absorption spectrometry (FI-AAS) system were studied using samples of less than 100 µl. The dispersion coefficient D, half steady-state time t_½ and half steady-state volume V_½ were used for evaluation. The transport conduits and sample loops in previously reported manifolds were found to be the main sources of sample dispersion. The detector studied had a t_½ response time of 0.20 s, which was almost independent of the sample introduction rate. Dispersion effects in the nebulizer-burner system, which influence the response kinetics of the detector, generally showed a very limited contribution of the over-all dispersion of the injected samples. Knotted sample loops, and short transport conduits of 0.35 mm i.d., were used in an improved Fl sample introduction manifold in order to limit dispersion. By injection 45 and 65 µl samples, 90 and 98%, respectively, of the steady-state signal from conventional sample aspiration were achieved. The precision and detection limits obtained by peak height measurement at an optimized carrier flow-rate of 4.2 ml min^–1 were comparable to conventional methods and were better than peak area measurements by a factor of 2. An absolute detection limit of 3 ng was obtained for lead in the peak height mode with a 35 µl sample, compared with 80 ng by conventional steady-state measurements. Tolerance of a high content of dissolved solids in samples was improved and demonstrated by repeated determinations (65 µl injections) of 10 µg ml^–1 of Pb in saturated lithium borate solution at 360 samples h^–1 for 1 h with 0.9% relative standard deviation (RSD), and in 30% m/v NaCl solution at 180 samples h^–1 with 2% RSD.