Phosphate adsorption in oxidized marine sediments

Abstract

The oxidized surface layer of marine sediments can act as a trap for phosphate diffusing upwards from deeper layers, thus influencing the flux of phosphate to the overlying water. Under oxic conditions adsorption of phosphate on metal (hydr-)oxides is the major process responsible for binding of phosphate at short time scales. Quantitative insight in this process is essential for the understanding of phosphorus diagenesis in sediments. Numerous studies on the interaction of phosphate with natural and synthetic Feand AI(hydr-)oxides (Parfitt, 1989; Torrent et al., 1992), soils (Parfitt, 1977) lake sediments (Jensen et al., 1992), and marine sediments (Krom and Berner, 1980; Sundby et al., 1992 ) have been performed under varying experimental conditions. In most cases, however, the initial concentrations used during the sorption experiments are much higher than the prevailing pore water concentrations in oxidized sediments or the results at these low concentrations are not reliable due to experimental difficulties and/or a limited number of data points. The aim of this study was to quantify the adsorption of phosphate on oxidized surface sediments with different sediment properties at concentrations comparable to natural levels found in the pore waters. At 8 locations in the North Sea (INP/BELS program; February 1992) sediment from two box cores (d-31 cm ) was subsampled with 20 acrylic liners (d= 5.6 cm ). All subsequent procedures took place at in situ temperature ( ~ 6 ° C). The upper 05 mm of the sediment from all cores was pooled and experiments started within a few hours. Characteristics of the sediments are given in Table 1. Sediment was incubated in 50 ml centrifuge tubes with filtered (0.2/Lm) North Atlantic surface water containing low nutrient concentrations ( [PO4]=0.05 /tmol 1-1) spiked with varying amounts of dissolved phosphate. All tubes were immediately shaken vigorously and then periodically (every 15 min during the first hour, subsequently every few hours) until sampling after 48 hr (isotherms) or shorter periods (kinetics). After centrifugation (1700 g; 15 min) the supernatant was removed, filtered (0.45 pm) and analyzed for Si and PO4 (Strickland and Parsons, 1972). Sediment pore water prior to incubation was obtained by the method of Saager et al. (1990). The amount of phosphate adsorbed on or desorbed from the sediment was calculated from the change in dissolved phosphate during the experiment. Results from the kinetics experiments at varying initial PO4 concentrations and at two solid/solution ratios, as shown for station 5 in Fig. 1, indicate that phosphate is both adsorbed and desorbed rapidly by the sediments. Approximately 80% of the total amount adsorbed after 48 hr is adsorbed within 15 min when [POa]initial=7.05 /tmol 1-1 and sediment /solut ion= 1:1 (Fig. 1A). At all initial