Abstract: Four extraction methods described by: (i) Psenner et al. (Arch. Hydrobiol., 1984, 111); (ii) Fitzgerald and Nelson (J. Phycol., 1966, 2, 32-37); (iii) Clark et al. (J. Bact., 1986, 168, 1212); and (iv) Mino et al. (Water Sci. Technol., 1985, 17, 11) were compared using a homogenized sample of a pure culture. FIA was used to determine total phosphate in the unfiltered sample, after digestion with K2S2O8. A HPLC system with an online detection capability was used to analyze phosphorus species (cf. Z. Wass, Abwass. Forsch., 1992, 25, 353). P1-P4 phosphates were separated on a LCA AO3 anion-exchange column (1.25 m x 4.6 mm i.d.). High mol. wt. phosphates were separated on a 5 µm LiChrospher 250 diol gel column (2.5 m x 1 cm i.d.). The eluents in both separations were EDTA/KCl. The polyphosphates were hydrolyzed to orthophosphates with HNO3 at 140°C. Then it is photometrically determined as yellow phosphorus vanadomolybdenium acid at 405 nm. Method (iv) gave the best results, 30% of the total phosphate was present as polyphosphate.

Abstract: A continuous-flow hydride generation AAS system (suitable for monitoring water supplies) is described. In order to determine As(III) and As(V) with the same sensitivity the sample is continuously mixed with concentrated HCl and a KI - ascorbic acid solution in a 1.5-m reaction coil heated to 75°C. The calibration graph extends from 1 to 100 µg L-1 with a rectilinear range up to 20 µg L-1 and a detection limit of 0.6 µg l-1. The chosen reaction medium for the selective determination of As(III) was 0.5 M acetic acid which gave a working range from 1 to 50 µg L-1 with a rectilinear range up to 20 µg l-1. With this determination Cu, Fe and Ni can interfere at a concentration. of 1 mg l-1, but the interferences can be suppressed by the addition of thiourea (60 mM) and EDTA (20 mM) to the acetic acid matrix.