Recovery of phosphorus, magnesium, and potassium from surplus activated sludge using existing sludge treatment facilities and thickened primary sludge

The accumulation of crystalline struvite causes the blockage of sludge transport pipes and dewatering screw-press screens. Therefore, both for the nutrient recycling and prevention of struvite accumulation, the recovery of P, Mg, and K from SAS is becoming critical.

The addition of acetate to SAS and subsequent anaerobic cultivation in the laboratory promoted the release of P, Mg, and K from the sludge solid (Wang et al. 2016; Ito et al. 2017; Salehi et al. 2018). Shiraiwa et al. (2018) used thickened primary sludge (TPS) as an organic carbon source, successfully released P, Mg, and K from SAS solids anaerobically using an existing storage tank, and recovered a separate liquid (SL) containing the released elements using an existing screw-press thickener at a full-scale sewage treatment plant (STP). They found that the addition of TPS to SAS in the storage tank accelerated the anaerobic release of the elements from the SAS solid. Therefore, the availability of Mg for the formation of struvite in the anaerobic digester decreased. Further, the SL with a comparatively high content of dissolved elements could be used as a raw material to produce liquid or solid fertilizers. The release rate of the elements was affected by the operating conditions; however, optimal operating conditions were not identified.

The present study examined how the anaerobic release of P, K, and Mg from SAS solids and their transfer to SL is affected by various operating conditions, including the amount of TPS added to SAS, oxidation-reduction potential (ORP) values of SAS and mixed sludge (i.e. mixture of TPS and SAS), total solid (TS) concentration in SAS, and water temperature of SAS. We used regression analysis to determine the operating conditions for optimal anaerobic release. Furthermore, we examined the recovery of P, Mg, and K from SL, through precipitation.

Tonan STP, located in the upper stream of the Kitakami River in Iwate Prefecture, Japan, was selected for the present study. At Tonan STP, sewage with an average flow rate of 123,398 m³/day is treated through a conventional activated sludge process, where an anaerobic zone is installed in front of the aeration tank to prevent bulking (Figure 1). PS and SAS are thickened with gravitational and mechanical thickeners, respectively; and TPS and thickened SAS (TSAS) are mixed and anaerobically digested together. For the purpose of this study, we assumed that PAOs are included in the SAS. Tonan STP is equipped with a sludge transport pipe and pump that in case of an emergency carries TPS to the SAS storage tank (see Figure 1). For this study, we used the sludge transport pipe to add TPS to the SAS storage tank (volume, 66.6 m³), thus accelerating the anaerobic release of P, Mg, and K from SAS.

First, we stopped the supply of SAS to the storage tank and subsequent mechanical screw thickening. Then, anaerobic cultivation was started in batch mode by adding TPS to the storage tank, where the volume of SAS was adjusted to 28.5 m³ (water depth, 2.3 m). The amounts of TPS added to the tank were 1.5 and 3 m³ in summer and 1.5, 3, and 6 m³ in winter. No addition of TPS was examined in both seasons as a control. The operating conditions with 1.5 m³ of TPS in summer and 3 m³ in winter were implemented twice on different days to ensure consistency. The TPS and SAS mixture and SAS alone were agitated for 30 s to mix the liquid and solid phases immediately after the addition of TPS and after 0.75, 1.5, 2.25, 3, 4, 5, and 6 h from the start of anaerobic cultivation. The cultivated time was set to 6 h totally, because the retention time of SAS in the existing storage tank is several hours.

To analyze the concentrations of elements, samples were collected in plastic bottles and glassware, preliminarily washed with 10% nitric acid (Kanto Chemical Co., Inc., Japan), and rinsed with super pure water (PURELAB flex-UV, ORGANO, Japan). Sludge samples were collected from the drainpipe at the bottom of the storage tank before and after the addition of TPS. Then, mixed sludge samples were collected 1.5, 3, and 6 h after the start of anaerobic cultivation. A sample of TPS was also collected before its addition to the storage tank. In addition, TSAS and SL samples were collected from the screw thickener before and after anaerobic cultivation.

The ORP, pH, dissolved oxygen (DO), and water temperature of the collected sludge samples were immediately measured using a portable ion・pH meter (type IM-32P, Toa DKK, Japan), portable dissolved oxygen meter (DO-31P, Toa DKK, Japan), and thermometer. The water temperatures of the pure SAS samples ranged from 18 to 25 °C in summer and from 14 to 15 °C in winter.

The collected samples of SAS, TPS, and SL were centrifuged at 3,000 rpm in the Tonan STP laboratory to obtain the liquid phase of the samples. The resultant supernatant was filtered using a membrane filter with a pore size of 0.45 μm (A045A047A, ADVANTEC, Japan). The filtrate was separated into two plastic bottles to measure the concentrations of soluble elements and dissolved organic carbon (DOC). Further, nitric acid was added to prevent the precipitation of released elements. The samples were transferred to our laboratory and stored in the refrigerator at 5 °C prior to pretreatment or analysis.

The samples were pretreated and analyzed according to the wastewater examination method (Japan Sewage Works Association 2012). For the measurement of the concentrations of total and dissolved elements (Ca, K, Mg, and P), first, to decompose the samples, we added hydrochloric and nitric acids (Hayashi Pure Chemical Ind., Ltd, Japan) and heated the samples. The digested samples were then filtered with the membrane filter with a pore size of 0.45 μm (A045A047A, ADVANTEC, Japan).

The concentrations of elements in the filtrate were measured using an inductively coupled plasma optical emission spectrophotometer (ICPE-9000, Shimadzu, Japan). A multielement standard solution (111355, Merck, Germany) was used to prepare standard solutions for the quantification of the elements. The DOC concentration in the filtrate without pretreatment was measured using a total organic carbon analyzer (TOC-V CSH, Shimadzu, Japan). For the measurement of total and volatile solids (TS and VS), the sludge samples were dried using a dry heat sterilizer (SSR-115, Isuzu, Japan) at 105 °C for 12 h, and the dried samples were then heated using an electric furnace (EPDW-7.2R Isuzu, Japan) at 600 °C for 6 h.

We then performed a single linear regression analysis on R software, to estimate the relationship between the release efficiency (RE) of P (dependent variable) and operational conditions, including the addition of TPS, water temperature, ORP, and TS (independent variables). We used the ‘lm’ function in R (R Core Team 2021). The analytical samples were prepared in triplicate and their average values are reported.

To recover the elements from SL, the precipitation of elements was examined using NaOH and/or CaCl₂ solutions. The analyzed SL sample was collected from the screw thickener after anaerobic cultivation, when 3 m³ of TPS was added to SAS during the summer season. SL was filtered through a qualitative filter paper with a pore size of 6 μm (000011600, ADVANTEC, Japan), and the filtrate was placed in glass beakers.

A 1 mol/L CaCl₂ solution was added to the filtrate (0.5 L) so that the molar ratio of Ca to P ([Ca]/[P]) became 2, and 0.1 and 1 mol/L NaOH solutions were added to the filtrate to adjust the pH within the range from 7.5 to 9. Furthermore, 1 mol/L CaCl₂ solution was added to 0.2 L of filtrate to adjust the molar ratio of Ca to P to 0.5, 1, 1.5, and 2, and the pH was adjusted to 9 by adding 0.1 and 1 mol/L NaOH solutions. The mixtures were left to stand for 30 min after stirring for 1 h, and the resulting supernatants were collected from the beakers.

The concentrations of the elements (Ca, K, Mg, and P) in the supernatants were measured using the same method described in section 2.1. The precipitate (1.5 m³, 14 °C) collected by centrifuging the mixture was dried and then analyzed using an X-ray diffractometer (XRD, Rint 2,200 V, Rigaku, Japan). The XRD pattern of the precipitate was compared with that of a reference in a database (PDXL, Rigaku, Japan). The analytical samples were prepared in triplicate and their average values are reported.

Table 1 shows the TS and total element concentrations in the SAS and TPS samples as well as the water temperatures of the SAS samples. The Tonan STP operates at higher mixed liquor-suspended solids concentrations in winter because of the decreased activity of activated sludge microorganisms at low temperatures. Therefore, TS concentrations in SAS at water temperatures of 13–15 °C were frequently higher than those in summer with warmer water temperatures of 18–24 °C. The concentrations of P, K, and Mg in the SAS samples ranged from 152 to 422 mg/L, 34.4 to 101 mg/L, and 36.8 to 76.4 mg/L, respectively. The total concentrations of P, K, and Mg in TPS ranged from 141 to 363 mg/L, 27.2 to 169 mg/L, and 22.2 to 71.1 mg/L, respectively.

Further, the concentration of K did not decrease at pH = 9. The optimal pH condition is approximately 11 for the precipitation of MgKPO₄ from synthetic wastewater containing P, K, and Mg (Xu et al. 2011; Nakao et al. 2017; Nagare et al. 2020). However, we did not adjust the pH to 11 because this would require more alkali salt for precipitation, and more acid for subsequent neutralization.

Ion activity (e.g. {Mg²⁺}) is expressed as follows (Xu et al. 2015):

Anaerobic cultivation of SAS generated from anaerobic and oxic processes released P, Mg, and K from SAS microorganisms such as PAOs in the existing storage tank before thickening of SAS. The release of P, Mg, and K was accelerated by adding TPS to SAS in the storage tank, which would be due to the uptake of dissolved organic matters supplied from TPS by PAOs and the promoted degradation of polyphosphates including K and Mg as counter cations inside PAOs. About 60% of P, Mg, and K in SAS were transferred to SL using the existing mechanical thickener. The transfer of P and Mg to the SL results in a decrease in P and Mg loadings to the subsequent anaerobic digester of sludge for inhibition of struvite formation. However, a method other than the precipitation process would be needed for the simultaneous recovery of P and K resources from SL. Electrodialysis is a promising technology for the recovery of P, K, and Mg ions from SL (Shiraiwa et al. 2018; Ward et al. 2018; Ye et al. 2019), for further use as fertilizers in agriculture.

When the method proposed in this study is actually operated, TPS, the existing storage tank before SAS thickening and mechanical thickener can be used as an alternative organic carbon source and for the release of P, K, and Mg from SAS and the subsequent separation of the released P, K, and Mg from SAS, respectively. However, pump and transport pipe would be needed for transporting TPS from the gravity thickener to the storage tank as shown in Figure 1. Furthermore, simultaneous recovery of P and K from SL would require additional facilities such as a filterer for removing suspended solids from SL, an electrodialyzer, and a storage tank for concentrated liquid. In future work, economic feasibility will need to be examined in detail for the scale-up of the proposed method.

P, Mg, and K were efficiently released from the SAS solid through anaerobic cultivation over 6 h after the addition of TPS to the SAS in an existing storage tank. Released elements were transferred from SAS to SL using an existing mechanical thickener. The addition of TPS and increased TS concentration in SAS increased the concentrations of P, K, and Mg in SL after anaerobic cultivation to 200, 58, and 37 mg/L, respectively. These findings suggest that TPS at existing STPs could be used to recover a liquid rich in valuable nutrients. Moreover, the recovery of elements lowers the Mg content in the anaerobic digester and inhibits struvite formation. P in SL was recovered as a hydroxyapatite precipitate by adding CaCl₂ and NaOH solutions. The optimal pH and molar ratio of Ca to P were 9 and 1, respectively. However, Mg and K could not be precipitated together with P. Further investigation is needed to develop a process for the simultaneous recovery of K, Mg, and P from SL.

This study was supported financially by the Iwate University; the Ministry of Land, Infrastructure, Transportation and Tourism of Japan; and the Fuso Innovative Technology Fund.

Yudai Kamiyama developed the methodology, investigated the work, rendered support in data curation and wrote the original draft of the paper. Takuya Shiraiwa developed the methodology and investigated the work. Nao Ishikawa developed the methodology, investigated the work, validated the data, and reviewed the article. Shinji Takahashi rendered support in data curation, and wrote the original draft of the paper, Makoto Sasamoto developed the methodology and investigated the work. Masaaki Sasaki developed the methodology and investigated the work. Toru Watanabe developed the methodology, investigated the work, and reviewed the work. Ayumi Ito conceptualized the whole article, investigated the work, validated the data, wrote the review and edited the article.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.