Physicochemical properties related to long-term phosphorus retention by drinking-water treatment residuals

Abstract

Drinking-water treatment residuals (WTRs) are nonhazardous materials that can be obtained free-of-charge from drinking-water treatment plants to reduce soluble phosphorus (P) concentrations in poorly P sorbing soils. Phosphorus sorption capacities of WTRs can vary 1-2 orders of magnitude, on the basis of short-term equilibration times (up to 7 d), but studies dealing with long-term (weeks to months) P retention by WTRs are lacking. Properties that most affect long-term P sorption capacities are pertinent to the efficacy of WTRs as amendments to stabilize P in soils. This research addressed the long-term (up to 80 d) P sorption/desorption characteristics and kinetics for seven WTRs, including the influence of specific surface area (SSA), porosity, and total C content on the overall magnitude of P sorption by seven WTRs. The data confirm a strong but variable affinity for P by WTRs. Aluminum-based WTRs tended to have higher P sorption capacity than Fe-based WTRs. Phosphorus sorption with time was biphasic in nature for most samples and best fit to a second-order rate model. The P sorption rate dependency was strongly correlated with a hysteretic P desorption, consistent with kinetic limitations on P desorption from micropores. Oxalate-extractable Al + Fe concentrations of the WTRs did not effectively explain long-term (80 d) P sorption capacities of the WTRs. Micropore (CO 2-based) SSAs were greater than BET-N2 SSAs for most WTRs, except those with the lowest (<80 g kg-1) total C content. There was a significant negative linear correlation between the total C content and the CO2/N2 SSA ratio. The data suggest that C in WTRs increases microporosity, but reduces P sorption per unit pore volume or surface area. Hence, variability in C content confounds direct relations among SSA, porosity, and P sorption. Total C, N2-based SSA, and CO 2-based SSAs explained 82% of the variability in the long-term P sorption capacities of the WTRs. Prediction of long-term P sorption capacities for different WTRs may be achieved by taking into account the three proposed variables.