Two manifold designs were evaluated. Water samples and wine digests in 10% nitric acid were pumped through a column containing a commercially available resin (Pb-Spec ® ), an immobilized crown ether with a cavity size selective for Pb 2+ . The column was rinsed with 2% HNO 3 and the eluent, 0.1 mol l −1 ammonium oxalate was injected via a six-port rotary valve. The eluted lead was delivered to the flame atomic absorption spectrometer at 4.0 ml min −1 . The following flow-injection (FI) parameters were optimized: sample acidity and volume, loading and elution flow rates, and eluent composition and volume.The detection limit for the water samples, estimated from the noise on the signal obtained for 250 ml of 10 g l −1 loaded at 19.1 ml min −1 was 1 g l −1 . For 50 ml of wine digest loaded at 4 ml min −1 , the value was 3 g l −1 . The roles of loading flow rate and sample volume were investigated in detail. The variation in retention efficiency with loading flow rate showed that the amount of lead retained (during a fixed loading time) increased with flow rate until the upper performance limit of the peristaltic pump was reached. The variation of detection limit with sample volume followed the expected hyperbolic relationship and showed that only small improvements in LOD were obtained for volumes greater than 50 ml. The method was evaluated through spike recovery for both water and wine. The lead contents of tap (0.24 g l −1 ), pond (0.40 g l −1 ), and river waters (not detected) were determined. The concentrations of lead in three Port wine samples ranged from not detected to 190 g l −1 . No significant matrix suppression effects were observed.

No abstract

No abstract

No abstract

No abstract