Calcium phosphate precipitation by seawater-driven forward osmosis without chemical addition.

A phosphorus (P) recovery technology based on forward osmosis (FO), driven by seawater, without any chemical addition, is tested at laboratory scale on sewage sludge digestate dewatering liquor.

The proposed method targets (and is thus limited by) dissolved P in sludge centrates from anaerobic digestion of sewage sludge. Thus, it targets mainly sewage treatment plants applying enhanced biological phosphate removal in combination with sludge digestion. Typically only these plants have sufficiently high soluble P levels in the sludge to allow economical recovery. In contrast to current struvite recovery processes, the proposed FO method does not require any chemical addition (i.e. no magnesium addition) and leads to calcium phosphate precipitate. The use of seawater as a draw solution has the advantage that regeneration of the draw solution is not required.

According to the authors forward osmosis has several advantages compared to other membrane technologies: (I) up to 97 % of initial P can be retained in the concentrated centrate (II) fouling can easily made reversible by washing the membranes with pure water (III) bidirectional flow leads to pH increase in the feed solution allowing calcium phosphate precipitation (IV) seawater can be used as a draw solution and thus as a source for Ca.

A lab scale cross flow FO system was used with a membrane surface are of 123 cm2 and two flow channels (CTA cellulose triacetate membranes). A sample of sewage sludge dewatering liquor (3 liter, filtered at 0,5 um, 29 mg P/l) was collected from a sewage treatment plant in New South Wales, Australia (configuration of the plant was not reported). This feed solution is separated by a membrane from the draw solution which is in this case was real seawater (5 liter, < 0.1 mg phosphate /L). These solutions were circulated at a rate of 1 l/min in the cross flow FO membrane system. Due to osmosis water flows from the feed to the draw solution leaving a more alkaline solution behind which is concentrated in Ca, Mg, K and P. Transfer of Ca or Mg from the draw solution to the feed solution was minimal, whereas K and Na showed transfer to the feed solution.

In this laboratory scale system it took around 3 days until 80 % of water volume moved from the feed solution to the draw solution but this can be expected to be considerably reduced in a full-scale system. The results indeed showed that P, Ca, Mg and K could be retained in the feed solution, most likely due to negative repulsion and/or size exclusion. At the same time the pH increased from 8.0 to 8.7. About 92 % of the initial P in the feed solution precipitated, about 4 % ended in the draw solution and about 4 % remained as dissolved P in the feed solution. Ca and P precipitated in significant quantities but also Mg, C, O and organic matter were found in the precipitates.

Mixed precipitate
The P content in the precipitate was only 3 % suggesting that next to calcium phosphate also calcium carbonate was formed. P-recovery was optimal at 65% water recovery and further concentration did not improve the recovery, probably due to a lack of remaining Ca and P in the solution. Over time the osmotic pressure decreased, thus reducing the membrane flux.

When the sludge centrate was three times concentrated the water flux went down to 30 % of the initial value. Draw solutions with higher ionic concentrations than seawater could improve P recovery further (e.g. desalination brine). Fouling of the membrane contributed only slightly to the decline in the water flux. In this experiment the fouling was fully reversible after washing the membrane with deionized water, this also shows that mineral formation took mainly place in the bulk solution. The authors suggest that the technology is interesting but that it requires economic evaluation (e.g. possible use of higher concentrated draw solution) and that the P content of the precipitate has to be increased.

"Phosphorus recovery from digested sludge centrate using seawater-driven forward osmosis." Separation and Purification Technology 163 (2016): 1-7

A. Ansari, F. Hai, L. Nghiem, Strategic Water Infrastructure Laboratory, School of Civil, Mining and Environmental Engineering, University of Wollongong, Wollongong, NSW 2522, Australia. W. Price, Strategic Water Infrastructure Laboratory, School of Chemistry, University of Wollongong, Wollongong, NSW 2522, Australia                

Return to the SCOPE In Press section.

This website uses cookies. They help us understand how visitors use our website so we can give you the best experience possible. By continuing to browse this site or choosing to close this message, you give consent for cookies to be used. Read more about cookies here.