Phosphorus recovery from incineration ash of phosphorus rich waste streams

Phosphorus (P) is an essential nutrient for all living organisms. Currently, P used in fertilizers and animal feed is largely extracted from non-renewable phosphate rock of which reserves are depleting. The European Union (EU) considers both P and phosphate rock as critical raw materials because of their strategic importance and the economic dependence on a limited number of non-EU suppliers. In this regard, it is important to recover P from alternative sources, in furtherance of a circular economy.

Biomass waste streams such as sewage sludge (SS) originating from wastewater treatment, animal manure, animal by-products and food waste contain large amounts of P. Currently, P is only partly recycled from these biomass waste streams, i.e., mainly through the use of animal manure as organic fertilizer. To reduce the demand for mineral P from non-renewable phosphate rock, it is imperative to further increase the P recycling/recovery rate from P rich biomass waste streams. To enable this process, it is preferable to first incinerate the biomass waste stream because this reduces its volume and mass, and concentrates the P in the obtained ash. Furthermore, incineration also recovers part of its energy content while destroying pathogens and toxic organic compounds the biomass waste stream might contain. However, both the low P bioavailability and high heavy metal concentrations in the obtained biomass ash impede its direct application as a P fertilizer in agriculture. Therefore, alternative ways to recover the valuable P from biomass ash have to be considered, such as wet chemical extraction, thermochemical treatment or electrodialysis. This research focused on wet chemical extraction because of its simplicity, high (energy) efficiency, low costs and the industrial precedents for P extraction from phosphate rock.

The overall aim of this research was to develop a highly efficient, cost-effective method for wet chemical P extraction from sewage sludge ash (SSA), poultry manure ash (PMA) as well as meat and bone meal ash (MBMA). The research consisted of three parts, the first of which involved the characterization of the biomass ash types for their elemental composition and ash mineralogy. In the second part, P extraction from the biomass ash types was optimized by studying different process variables and considering heavy metal co-extraction. The third part assessed whether incineration conditions affect the biomass ash mineralogy and related P and heavy metal extraction.

It was found that the investigated biomass ash types (SSA, PMA and MBMA) differ in composition and mineralogy. The SSA showed higher concentrations of Al, Fe, Si and heavy metals, whereas PMA and MBMA showed a higher Ca concentration. These variations in elemental composition were reflected in differences in P mineralogy, since it was found that approximately 40% of the P in SSA was bound to Al or Fe (Al/Fe-phosphates), while no Al/Fe-phosphates were found in PMA and MBMA. The latter mainly consisted of P bound to Ca or Mg (Ca/Mg-phosphates), which were found to bind approximately 60% of the P in the SSA. Differences in elemental composition and P mineralogy did not affect the P extraction efficiency with inorganic acids. In contrast, for the organic acids, alkaline extraction liquid and chelating agents considered, the P extraction efficiency was highly affected by the P mineralogy and elemental composition of the ash, and could be related to the chemical characteristics of the considered extraction liquids. Alkaline extraction liquids showed low heavy metal co-extraction when compared to acidic extraction liquids. The highest P extraction efficiencies (> 88%) for all biomass ash types were obtained for inorganic acid extraction, which consistently coincided with high heavy metal co-extraction. To conclude, it was found that the type of extraction liquid, ash composition and ash mineralogy all affected P and heavy metal extraction.

Further research specifically focused on P extraction from SSA. Design of experiments results showed that extraction liquid concentration (0.1 – 0.5 N), liquid/solid ratio (10 – 50 ml/g ash), contact time (10 – 120 min) and pH all affected P and heavy metal extraction from SSA. It was concluded that a P extraction efficiency > 85% could only be obtained at a pH < 2 since both Ca/Mg- and Al/Fe-phosphates in the SSA dissolved well at these low pH values. At slightly higher pH only Ca/Mg-phosphates dissolved well and at alkaline pH only Al/Fe-phosphates dissolved well. Multi-criteria techno-economic optimization allowed the establishment of optimal extraction conditions based on a trade-off between high P extraction, low heavy metal co-extraction and low operational costs. Results showed that: (1) sulfuric acid (0.5 N, 10 ml/g ash, 120 min) outperformed the other extraction liquids in terms of extraction liquid costs per kg P extracted; and (2) oxalic acid (0.5 N, 12.8 ml/g ash, 120 min) showed the lowest heavy metal co-extraction, in this way reducing the downstream processing costs for heavy metal removal. However, for both extraction liquids, a purification step was still necessary to remove all remaining heavy metals from the extract. It is for this reason that in the third part of this research, it was evaluated to which extent the SS incineration conditions affect the SSA mineralogy and how this further reflects on the P extraction and heavy metal co-extraction from this ash.

It was found that the SS incineration temperature (550 – 1100 °C) affected both the P extraction and heavy metal co-extraction. Chemical addition prior to incineration (i.e., Al, Ca, Fe, Mg and Na chlorides, sulfates and carbonates) mainly affected the heavy metal co-extraction. The P extraction efficiency for the acidic extraction liquids considered showed a parabolic trend as a function of the incineration temperature, with a maximum obtained around an incineration temperature of 850 °C. On the other hand, the P extraction efficiency with sodium hydroxide showed a decrease as a function of the incineration temperature, related to a decrease in Al/Fe-phosphates in the SSA. Furthermore, heavy metal co-extraction decreased with increasing incineration temperature due to immobilization of the heavy metals. Heavy metal co-extraction was lowest in the SSA sample obtained at an incineration temperature of 1000 °C (< 21% total heavy metals extracted). However, the P extraction from the SSA obtained at this incineration temperature was < 73% because of the incorporation of P into silicate melt agglomerates. It could be concluded that an incineration temperature in the range 800 – 850 °C offered the best trade-off between high P extraction and low heavy metal co-extraction. The co-extraction of heavy metals could be further reduced by the addition of aluminum chloride, iron(III) chloride, magnesium chloride or aluminum sulfate to the SS prior to incineration. This had minimal effect on the P extraction efficiency. Moreover, it was found that the addition of calcium chloride, sodium chloride or carbonates to the SS prior to incineration should be avoided because it decreased the P extraction efficiency.

This research indicates that it is technologically feasible to efficiently extract P from incineration ash of P rich biomass waste streams with minimal co-extraction of heavy metals. However, further research should investigate the purification of the obtained P extract (heavy metal removal) and the final precipitation of P compounds for specific applications. Moreover, to complement the results of this research, an extended study on the environmental impact and techno-economic feasibility of the P recovery process should be performed. This will be part of further research within the ChEMaRTS research group (KU Leuven).