Recovery of nutrients (N-P-K) from potassium-rich sludge anaerobic digestion side-streams by integration of a hybrid sorption-membrane ultrafiltration process: Use of powder reactive sorbents as nutrient carriers.

Here, an alternative nutrient (N-P-K) recovery route from potassium-rich sludge anaerobic digestion side-streams using powder reactive sorbents (PRSs) is presented. In the first step, the optimum PRS system was determined in batch experiments with mixtures of: a) a sodium zeolite (NaP1) to facilitate the NH4+ and K+ sorption; b) a Ca-zeolite (CaP1) to facilitate the removal of P by formation of Ca-phosphates (e.g., CaHPO4(s)), and c) caustic magnesia containing mixtures of MgO to facilitate the formation of Mg/NH4/PO4 minerals (e.g., struvite and magnesium phosphates). Evaluation of the continuous and simultaneous N-P-K removal with mixtures of PRSs was carried out using a hybrid sorption/filtration system with ultrafiltration (UF) hollow-fibre membranes. The dosing ratios of the PRS mixtures were optimised on the basis of the equilibrium and kinetic sorption data, and a PRS dose (<2-5gPRS/L) was selected to ensure the hydraulic performance of the system. Under such conditions, and with synthetic anaerobic side-stream removal capacities (qt) of 220±10mgN-NH4/g, 35±5mgP-PO4/g, and 8±2mgK/g, removal efficiencies of 32±3, 78±5, and 26±3% for ammonium, phosphate, and potassium, respectively, were obtained for the binary mixtures of NaP1/CaP1 zeolites. Contrary to the batch results, the use of tertiary mixtures of NaP1/CaP1/MgO only improved the K removal capacity and efficiency to 18±2mgK/g and 55±4%, respectively, while the phosphate removal capacity and efficiency remained unchanged (ca. 35±3mgP-PO4/g; 80±5%) and the ammonium capacity and efficiency were reduced to 185±12mgN-NH4/g and 20±2%, respectively, due to the competing Mg2+ ion effect. Nutrient removal trials with real anaerobic side-streams using binary mixtures of Na/Ca zeolites showed a reduction of both the hydraulic performance and the nutrient removal ratios due to the presence of dissolved organic matter. However, constant removal ratios of N, P, and K were recorded throughout the filtration experiments. The loaded PRSs exhibited suitable nutrient release rates and bioavailability as co-substrates for soil quality improvement. Chemical analyses detected the formation of Ca/P/O and Mg/N/P/O neo-minerals; however, the mineralogical data revealed only the formation of struvite, even when no magnesium oxide was used.