Abstract

Phosphorus crystallization-filtration (PCF) was devised as a novel tertiary process for phosphorus removal from domestic wastewater. The results obtained showed that during the PCF process, high pH and excessive calcium dosage conditions were required to obtain effluents with total phosphorus (T-P) and suspended solid (SS) concentrations below 0.2 and 10 mg/L, respectively, within 2 h of operation. Phosphorus was precipitated during the pre-treatment step, and thereafter it crystallized on the surface of the fixed seed material in the PCF reactor. Furthermore, the addition of Ca2+ resulted in phosphorus removal efficiencies >95%, and pH, residual Ca2+, filtration depth, and linear velocity were identified as the main design and operation parameters of the PCF process. Following the pilot-scale PCF process, the average concentrations of T-P, PO4-P, and SS in the effluent were 0.05, 0.04, and 1.1 mg/L, respectively, corresponding to removal efficiencies of 90.9, 86.5, and 79.7%, respectively. The investigation of the backwashing sludge characteristics of the PCF process using scanning electron microscopy (SEM), Fourier transform-infrared vacuum spectrometry (FT-IR), energy dispersive X-ray spectroscopy (EDS), and X-ray diffraction (XRD) analyses showed that owing to its high contents in calcite and hydrated phosphorus compounds, PCF sludge could be used as an alternative soil amendment resource.

HIGHLIGHTS

  • Phosphorus crystallization-filtration (PCF) was devised as a novel tertiary process for phosphorus removal from domestic wastewater.

  • The investigation of the backwashing sludge characteristics showed high contents in calcite and hydrated phosphorus compounds.

  • PCF sludge could be used as an alternative soil amendment resource.

INTRODUCTION

In Korea, cyanobacterial algal blooms represent one of the major issues associated with lakes and rivers during all summer seasons. The primary substance responsible for this excessive growth of aquatic organisms is phosphorus, originating from wastewater treatment plants (WWTPs) and croplands. Kim et al. (2007) demonstrated that with respect to algal growth, phosphorus is a more important nutrient than nitrogen. They also stated that in Korean lakes and rivers, algal growth was controlled by phosphorus loads. To prevent eutrophication and algal bloom, the management of the phosphorus content of domestic wastewater is essential before its discharge (Bashan & Bashan 2004; Bal Krishna et al. 2016). Thus, in Korea, the total phosphorus (T-P) content of domestic wastewater effluent has been regulated under 0.2 mg/L in water resource conservation areas.

In domestic WWTPs, after the preliminary biological treatment followed by the secondary sedimentation process, the tertiary phosphorus removal process is performed. The most extensively used tertiary phosphorus removal process is coagulation, which is performed alongside sedimentation, filtration, or flotation. However, owing to relatively high variations in pollutant loads and flow rates, such tertiary phosphorus removal processes cannot be integrated into most small-scale WWTPs that are used to treat less than 500 m3 of influent per day (Lee et al. 2004). Owing to the unstable water quality as well as the varying quantity of influents that is characteristic of WWTPs in small agricultural villages, controlling the dosage of the coagulant required remains challenging. Additionally, coagulation-based processes generate large amounts of sludge, and their phosphorus recovery performance is very poor.

Therefore, to overcome the limitations associated with coagulation-based processes, new phosphorus removal processes are required. Deliyanni et al. (2007), Berg et al. (2006), Kim et al. (2006), and Joko (1985) reported that hydroxyapatite (HAP, Ca5(PO4)3OH) crystallization is a plausible alternative for the removal and recovery of phosphorus from wastewater. HAP crystallization involves the removal of phosphorus from wastewater in the form of HAP; that is, HAP is grown and recovered on the surface of seed crystals via the crystallization of phosphate, calcium, and hydroxyl ions under relatively high pH conditions. HAP growth as well as the associated phosphorus removal efficiency are affected by several reaction conditions, including pH, temperature, and the concentrations of phosphate, calcium, and bicarbonate ions. The quality of the recovered sludge, including purity, morphology, and particle distribution, is also affected by the Ca/P ratio, temperature, pH, ion concentrations, mixing conditions, as well as the sedimentation time of the crystallization process (Al-Harahsheh et al. 2014; Cichy et al. 2019). The equation for reaction that brings about HAP growth is as follows (Nancollas 1968): 
formula
(1)

Furthermore, calcite (Song et al. 2001), sand (Momberg & Oellermann 1992), cow bone (Jang & Kang 2002), tobermorite (Moriyama et al. 2001, 2003), xonotlite (Chen et al. 2009), and porous calcium silicate hydrate (PCSH) (Guan et al. 2014) have been suggested as seed crystals for the crystallization process. Usually, seed crystals for HAP crystallization are calcium-based materials that are characterized by high surface areas. Based on laboratory-scale experiments, several studies have reported the use of calcite in phosphorus removal; thus, it is considered one of the most promising materials for use as HAP seed crystals (Ishikawa & Ichikuni 1981; Song et al. 2001, 2006; Liu et al. 2012).

To achieve high phosphorus removal efficiency within a short operation time, high pH conditions and excessive Ca2+ dosage are required (Momberg & Oellermann 1992; Chen et al. 2009; Menglin et al. 2016; Cichy et al. 2019). Most laboratory-scale phosphorus removal experiments with artificial wastewater using HAP crystallization have shown that phosphorus removal to concentrations below 1.0 mg/L (PO4-P) could not be achieved. In a study conducted by Momberg & Oellermann (1992), under high pH conditions, an effluent phosphate concentration below 1.0 mg/L was achieved; however, the phosphate concentration did not remain below this threshold owing to the use of a fluidized sand media as seed crystals. Conversely, Donnert & Salecker (1999) reported that calcite can be used as seed material to achieve a stable phosphorus removal efficiency in both pilot- and full-scale industrial WWTPs without replacement for several years. They found that the average PO4-P concentrations in the effluent from the pilot- and full-scale industrial WWTPs were 1.3 and 1.1 mg/L, respectively. Based on regulatory standards, to apply calcite as seed crystal in the tertiary removal of phosphorus from Korean domestic wastewater, it is required that phosphorus should be removed within a short operation time such that its concentration in the effluent is below 0.2 mg/L. Additionally, the removal of suspended solids (SS) is also necessary.

The phosphorus crystallization-filtration (PCF) process could generate sludge containing calcite and phosphorus compounds under conditions of high Ca2+ and dissolved carbonate concentrations. Calcite and limestone powder are commonly used as soil amendment to neutralize acidic soils (Tozsin et al. 2014). Additionally, it has been estimated that in 100 years, phosphate resources will be depleted given that phosphorus is greatly consumed in the form of agricultural fertilizers (Woods et al. 1999). However, studies on phosphorus recovery technologies utilizing Ca-P crystallization in secondary treated effluents are scarce owing to their low recovery potential (Cordell et al. 2011; Egle et al. 2015).

In this study, the PCF process is presented as a novel tertiary process for the removal of phosphorus from domestic wastewater. The PCF process was designed to achieve an effluent T-P concentration less than 0.2 mg/L and also remove SS to concentrations below 10 mg/L within 2 h of operation. In the PCF reactor, Ca2+ and pH conditions were controlled during the pre-treatment step, and subsequently, HAP crystals were precipitated and crystallized on the surface of fixed seed calcite sands, while allowing for sufficient contact time. The discharged sludge from the PCF reactor containing calcite and phosphorus compounds could then be recycled or upcycled as a valuable resource.

Thus, to simulate the PCF process, cylindrical column tests were performed with operation parameters, including pH, additional Ca2+ dosage, and linear velocity. The design and operational parameters of PCF are suggested from the results of column tests. Finally, the pilot-scale PCF process was conducted. Furthermore, the sludge characteristics of PCF were evaluated for application as an alternative soil amendment resource for acidic soil neutralization.

MATERIALS AND METHODS

Laboratory-scale cylindrical column tests

To simulate the PCF process, cylindrical column tests (Figure 1) were performed using the effluents from a secondary sedimentation basin. The column had a diameter of 5 cm, a filtration depth of 1.4 m, and a volume of 2 L. Samples were collected from the column at 0.2 m intervals for analysis. The average diameter of the limestone sand, which was packed into two serial columns, was 2.0–3.0 mm. The influent water had the following characteristics: pH, 6.4–7.2; Ca2+, 25.7–32.4 mg/L; T-P, 0.6–2.4 mg/L; and PO4-P, 0.3–1.5 mg. Table 1 shows the experimental conditions of the column tests. To supply Ca2+ and OH and obtain pH values between 10 and 11, the effluent of the secondary sedimentation basin was pre-treated using 20% of Ca(OH)2. After the addition of 50 mg/L of Ca2+ using a 25% CaCl2 solution, the secondary sedimentation basin effluent was filtered through a 2.8 m depth of two serial columns under 2.0, 5.0, and 10.0 m/h linear velocity conditions.

Table 1

Experimental conditions of column experiments

Test no.Condition
pHCa2+ addition (mg/L)Linear velocity (m/hr)
10.0 2.0 
10.0 50 2.0 
11.0 2.0 
10.0 5.0 
10.0 10.0 
Test no.Condition
pHCa2+ addition (mg/L)Linear velocity (m/hr)
10.0 2.0 
10.0 50 2.0 
11.0 2.0 
10.0 5.0 
10.0 10.0 
Figure 1

Schematic diagram of column experiments for the PCF process.

Figure 1

Schematic diagram of column experiments for the PCF process.

Pilot-scale operation

A pilot PCF process plant was constructed at Ilsan Wastewater Treatment Center as shown in Figure 2. The plant consisted of a pre-treatment basin (A), a PCF reactor (B), and a post-treatment basin (C). The pre-treatment and post-treatment basins both had volumes of 1.0 m3, and the PCF reactor had the following specifications: diameter, 1.2 m; depth, 3.0 m; volume, 3.5 m3; moreover, it was filled with limestone (2.0–3.0 mm diameter) as seed material.

Figure 2

Pilot PCF process plant in the Il-san Wastewater Treatment Center (50.4 m3/day).

Figure 2

Pilot PCF process plant in the Il-san Wastewater Treatment Center (50.4 m3/day).

The operating conditions of the pilot PCF plant from June 2018 to September 2018 are shown in Table 2. The pre-treatment pH was adjusted to 10.5 ± 0.3 using 20% Ca(OH)2. More Ca2+ was added to reach a residual value >80 mg/L using 25% CaCl2 solution. The operating linear velocity of the reactor was 1.9 m/h (i.e. a flow rate of 50.4 m3/day). The quality of the raw water that was fed into the reactor was as shown in Table 3.

Table 2

Operating conditions of PCF pilot plant

Operating conditionSpecification/value
Filtration depth 2.5 m 
Media Limestone sand (2.0–3.0 mm) 
Flow rate 50.4 m3/day (L.V. 1.9 m/h) 
pH 10.6 ± 0.3 (20% Ca(OH)2
Residual Ca2+ >80 mg/L (25% CaCl2
Operating conditionSpecification/value
Filtration depth 2.5 m 
Media Limestone sand (2.0–3.0 mm) 
Flow rate 50.4 m3/day (L.V. 1.9 m/h) 
pH 10.6 ± 0.3 (20% Ca(OH)2
Residual Ca2+ >80 mg/L (25% CaCl2
Table 3

Raw water quality of PCF pilot plant

Water quality parameterValue (average)
Water temperature (°C) 25.2–33.5 (29.4) 
pH 6.4–7.0 (6.8) 
T-alkalinity (mg/L as CaCO343.9–127.7 (84.0) 
T-P (mg/L) 0.23–1.09 (0.61) 
PO4-P (mg/L) 0.07–1.04 (0.43) 
Ca2+ (mg/L) 21.9–34.2 (26.0) 
Conductivity (μS/cm) 363–607 (543) 
SS (mg/L) 0–77 (13.2) 
Water quality parameterValue (average)
Water temperature (°C) 25.2–33.5 (29.4) 
pH 6.4–7.0 (6.8) 
T-alkalinity (mg/L as CaCO343.9–127.7 (84.0) 
T-P (mg/L) 0.23–1.09 (0.61) 
PO4-P (mg/L) 0.07–1.04 (0.43) 
Ca2+ (mg/L) 21.9–34.2 (26.0) 
Conductivity (μS/cm) 363–607 (543) 
SS (mg/L) 0–77 (13.2) 

Characterization of PCF sludge

To evaluate the performance of the recycling process, the precipitated and filtered sludge samples obtained after the PCF process were analyzed. The pH was controlled in the range 10.5–10.8, and more Ca2+ (100 mg/L) was added. The initial PO4-P concentrations in the experiments were 0.16, 2.0, 10.0, and 50.0 mg/L. Using a centrifugal separator, the sludge samples were separated from the effluent and dried at 50 °C.

Thereafter, scanning electron microscopy (SEM, S-4800, HITACHI), energy dispersive X-ray spectroscopy (EDS, EDX S-10, Oxford), X-ray diffraction (XRD, MiniFlex600, Rigaku), and Fourier transform-infrared vacuum spectrometry (FT-IR, VERTEX 80 V, Bruker) were employed to determine the chemical composition of the PCF sludge samples.

RESULTS AND DISCUSSION

Laboratory-scale cylindrical column experiments

During the pre-treatment step, Ca(OH)2 released Ca2+ and OH ions. As shown in Figure 3, pH values increased from 7.0 to 10.0, and then to 11.0. Owing to Ca(OH)2 supplementation, Ca2+ concentration increased to ∼30 mg/L at a pH of 10.0 and 50 mg/L at a pH of 11.0, as shown in Figure 4. For tests 2–5, Ca2+ concentration was increased to ∼50 mg/L via the addition of more CaCl2. Within a filtration depth of 0.4 m, there was a rapid decrease in both pH and Ca2+ concentration. Beyond 0.4 m, both quantities decreased generally, depending on the conditions associated with the different filtration depths. Under high pH conditions and low linear velocity, the decrease in pH was higher than that observed under lower pH or higher linear velocity conditions. This implied that a higher pH and a lower linear velocity favored the rapid consumption of OH and CO32−. Additionally, higher phosphorus removal and calcite regeneration were expected under these conditions than under low pH and high linear velocity conditions.

Figure 3

pH variation with filtration depth.

Figure 3

pH variation with filtration depth.

Figure 4

Ca2+ variation with filtration depth.

Figure 4

Ca2+ variation with filtration depth.

The variations of T-P and PO4-P with depth are shown in Figures 5 and 6, respectively. There were two steps for phosphorus removal. Firstly, approximately 30–60% of T-P and PO4-P removal efficiencies were achieved via precipitation under the high pH condition of the pre-treatment step. In the case of test 2, in which the initial Ca2+ dosage was higher than in the other tests, more phosphorus was removed during the pretreatment step than under other conditions.

Figure 5

Variation of (a) T-P and (b) phosphorus removal efficiencies with filtration depth.

Figure 5

Variation of (a) T-P and (b) phosphorus removal efficiencies with filtration depth.

Figure 6

Variation of (a) PO4-P and (b) phosphorus removal efficiencies with filtration depth.

Figure 6

Variation of (a) PO4-P and (b) phosphorus removal efficiencies with filtration depth.

Secondly, phosphorus was gradually removed via crystallization and filtration as the filtration depth increased under all conditions. Without additional Ca2+ dosage, phosphorus removal efficiencies were approximately 75% at the end of the column. However, both T-P and PO4-P removal efficiencies remarkably increased and reached >95% under all pH conditions when the linear velocity ranged between 2.0 and 10.0 m/h owing to Ca2+ supplementation. In particular, phosphorus was rapidly removed within 0.8 m of filtration depth under supplemental Ca2+ conditions. The removal efficiency of phosphorus at a pH of 11.0 was not higher than that at a pH of 10.0. The highest T-P and PO4-P removal efficiencies (98.1 and 100.0%, respectively) were realized when the supplemental Ca2+ ion concentration was 50 mg/L, under a linear velocity of 2.0 m/h. In principle, under pH values above 10, coupled with additional calcium adjustment and a filtration depth above 2.0 m, phosphorus removal efficiency was stable.

A stable effluent phosphorus concentration below 0.2 mg/L was realized compared with the value that was obtained by Momberg & Oellermann (1992). The estimated total operation time of the PCF process was 1–2 h (40 min for pre-treatment and 20–80 min for crystallization-filtration), which was significantly shorter than those reported by Chen et al. (2009) (>24 h) and Momberg & Oellermann (1992) (>4 h).

Design and operating parameters for the PCF process

The schematic illustration, design parameters, and operational conditions of the PCF process for 50.0–100.0 m3/day of treatment capacity are shown in Figure 7 and Table 4.

Table 4

PCF design parameters and operational conditions

ParameterDesign parameterSpecification/value
Pre-treatment pH >pH 10.0 
Residual Ca2+ >80 mg/L 
OH supplier Ca(OH)2 
Ca2+ supplier Ca(OH)2, CaCl2 
PCF reactor Filtration media 2.0–3.0 mm limestone sand 
Filtration direction Up-stream 
Filter depth 2.0–3.0 m (depending on T-P load and linear velocity) 
Backwashing method Air lifting (intermittently operated according to pressure of PCF reactor) 
Linear velocity 1.0–5.0 m/h 
EBCT 40–180 minutes 
ParameterDesign parameterSpecification/value
Pre-treatment pH >pH 10.0 
Residual Ca2+ >80 mg/L 
OH supplier Ca(OH)2 
Ca2+ supplier Ca(OH)2, CaCl2 
PCF reactor Filtration media 2.0–3.0 mm limestone sand 
Filtration direction Up-stream 
Filter depth 2.0–3.0 m (depending on T-P load and linear velocity) 
Backwashing method Air lifting (intermittently operated according to pressure of PCF reactor) 
Linear velocity 1.0–5.0 m/h 
EBCT 40–180 minutes 
Figure 7

Schematic representation of the PCF process.

Figure 7

Schematic representation of the PCF process.

The T-P concentration in the influent was 1.0–3.0 mg/L, and it was expected that after the PCF process, it would be less than 0.2 mg/L in the resulting effluent. Based on the results of column experiments, T-P removal efficiencies were expected to reach >95%, and be stable at a 2.4 m filtration depth with a linear velocity of 2 m/h, accompanied with the supplementation of Ca2+. When the influent T-P concentration was less than 2.0 mg/L, a linear velocity in the range 2–10 m/h or a filtration depth <2.4 m allowed the removal of T-P to values below 0.2 mg/L. Thus, depending on the T-P content of the influent and linear velocity, the reactor could be designed to have a filtration depth in the range of 2.0–3.0 m. A linear velocity between 1.0 and 5.0 m/h, which represents an empty bed contact time (EBCT) in the range 40–180 min (20–90 minutes of hydraulic retention time (HRT) in 0.5 filter media porosity) in a cylindrical basin with a diameter of 1.2 m is recommended. Additionally, the PCF reactor should be filled with limestone sand (diameter, 2.0–3.0 mm).

To retain a sufficient contact time with the seed crystals and ensure stable phosphorus removal, fixed bed filtration was preferred over fluidized bed filtration. However, backwashing is necessary for fixed bed filtration when the filter bed becomes blocked.

The pressure in the PCF reactor could be measured and used as an operating index for the determination of the backwashing phase. The backwashing sludge containing phosphorus compounds, and calcite should be separated from the PCF reactor for recycling. The air lifting method can also be recommended for the backwashing of the media. This method probably resulted in the effective removal of the filtered sludge in the lower part of the reactor, leading to a recovery of the PCF reactor pressure.

Pilot-scale operation

The operation of the pilot-scale PCF process (June, 2018–September, 2018) was monitored, and different parameters were measured. The influent and operational pH values were in the ranges 6.4–7.0 (mean = 6.8) and 10.2–10.8 (mean = 10.5), respectively, as shown in Figure 8.

Figure 8

pH variation during the pilot-scale PCF process.

Figure 8

pH variation during the pilot-scale PCF process.

The T-P and PO4-P concentrations in the influent were in the range 0.23–1.09 (mean = 0.62) and 0.07–1.04 (mean = 0.43) mg/L, respectively, as shown in Figure 9. Despite the variation of the T-P concentration in the influent, the T-P in the effluent was in the range 0.03–0.11 (mean = 0.05) mg/L, indicating a 90.0% removal efficiency. Additionally, in the effluent, PO4-P concentration ranged between 0 and 0.06 (mean = 0.04 mg/L), indicating an average removal efficiency of 86.5%. During the monitoring period, it was observed that T-P and PO4-P concentrations were stably maintained below 0.2 mg/L.

Figure 9

Phosphorus removal during the pilot-scale PCF process.

Figure 9

Phosphorus removal during the pilot-scale PCF process.

Moreover, SS removal was also monitored, and the results are shown in Figure 10. The standard for effluent SS content is <10.0 mg/L. The influent SS content fluctuated between 0 and 77 mg/L. However, in the effluent, it was averagely 1.1 mg/L, indicating a 79.7% removal efficiency.

Figure 10

Suspended solids removal during the pilot-scale PCF process.

Figure 10

Suspended solids removal during the pilot-scale PCF process.

Interestingly, the influent PO4-P proportion of T-P varied in the range 23.6–100% (mean = 67.9%), suggesting the presence of both PO4-P (soluble) and particulate phosphorus (T-P – PO4-P) in the influent. The fractions of soluble and particulate phosphorus in real domestic wastewater could vary greatly. Regarding the lowest influent PO4-P ratio of T-P (8/02), crystallization-filtration resulted in successful T-P (PO4-P and particulate phosphorus) removal (96%). Thus, it was demonstrated that the PCF process can be adapted to suit a wide variety of soluble and particulate phosphorus fractions in domestic wastewater. Compared with the results reported by Donnert & Salecker (1999) based on pilot-scale experiments using calcite, the PCF process in this study resulted in a lower and relatively stable effluent PO4-P content.

Characterization of the PCF sludge

The sludge obtained after the PCF process was analyzed and evaluated as an alternative soil amendment. Figure 11 shows the SEM images of the PCF sludge obtained from influent with adjusted PO4-P concentrations in the range 0.16–50.0 mg/L. The sludge resulting from the influent with initial PO4-P content <0.2 mg/L was characterized by an amorphous layer without any particles (Figure 11(a)). However, particles with diameter ∼30 nm or less appeared on the surface of the amorphous layer of the sludge obtained from the influent with an initial PO4-P concentration of 2.0 mg/L (Figure 11(b)). Additionally, the diameter of sludge particles resulting from influents with initial PO4-P concentrations of 10.0 and 50.0 mg/L were greater (Figure 11(c) and 11(d)) relative to the particles shown in Figure 11(b). Notably, the sludge resulting from the influents with higher initial PO4-P concentrations did not show any amorphous layers. They consisted of accumulated particles with diameters of 50 nm or larger.

Figure 11

SEM images of PCF sludge obtained from influents with initial PO4-P concentrations of (a) 0.16, (b) 2.0, (c) 10.0, and (d) 50.0 mg/L.

Figure 11

SEM images of PCF sludge obtained from influents with initial PO4-P concentrations of (a) 0.16, (b) 2.0, (c) 10.0, and (d) 50.0 mg/L.

Calcite and phosphorus enrichment in the PCF sludge was interpreted based on FT-IR peaks. Figure 12 shows the FT-IR diagrams of the sludge samples based on the initial PO4-P values. Typical wave numbers that indicate the presence of PO43− include 560–600 and 1,000–1,100 /cm. Additionally, a wave number of 875 /cm indicates the presence of HPO42−, while a wave number of 873 /cm and ∼1,420 /cm indicates the presence of CO32−. Wave numbers in the range 2,600–3,600 /cm indicate the presence of H2O (adsorbed water). The absorbance of the sludge samples obtained from influents with lower initial PO4-P conditions presented larger peaks at wave numbers corresponding to CO32−, possibly implying that in the sludge, calcite (CaCO3) fractions were higher than phosphorus compound fractions. On the other hand, the absorbance of the sludge samples obtained from influents with higher initial PO4-P contents showed larger peaks at wave numbers corresponding to PO43− and H2O, implying that the sludge compounds were hydrated under conditions of higher initial PO4-P content.

Figure 12

FT-IR diagrams for PCF sludge obtained from influents with initial PO4-P concentrations of 0.16, 2.0, 10.0, and 50.0 mg/L.

Figure 12

FT-IR diagrams for PCF sludge obtained from influents with initial PO4-P concentrations of 0.16, 2.0, 10.0, and 50.0 mg/L.

Additionally, based on EDS analysis, the average atomic ratios of the phosphorus particles obtained from influents with initial PO4-P concentrations in the range 0.16, 2.0, 10.0, and 50.0 mg/L were 0, 1.3, 2.7, and 9.5%, respectively. FT-IR and EDS results also confirmed that sludge from influents with higher initial PO4-P concentrations contained more hydrated phosphorus compounds than calcite.

The main crystal composition of the PCF sludge was determined using XRD analyses (Figure 13). Peaks corresponding to calcite were the most predominant, while peaks corresponding to HAP crystal were insignificant. Synthetic carbonate-hydroxylapatite (Ca10(PO4)3(CO3)3(OH)2) and phosphoric acid (H3PO4) were also detected; however, their peaks were also insignificant. EDS analysis showed an increase in phosphorus atomic ratios, and FT-IR showed significant absorbance peaks corresponding to phosphorus compounds depending on the initial PO4-P concentration. However, a few minor peaks corresponding to HAP crystal peaks were detected. The phosphorus removed during the PCF process was assumed to be formed as a major portion of amorphous hydrated phosphorus compounds and a minor portion of the HAP crystal.

Figure 13

XRD analysis results of the PCF sludge.

Figure 13

XRD analysis results of the PCF sludge.

Overall, the results of this study showed that calcite and hydrated phosphorus compounds are the main components of backwashing sludge. This conclusion was inferred from FT-IR and XRD data. Additionally, the fact that the phosphorus portion of the recycled sludge could be larger with higher initial PO4-P concentrations was confirmed by FT-IR and EDS data. Thus, when attempting to recycle the backwashing sludge of PCF as an alternative soil amendment or fertilizer, it should be noted that the phosphorus content and composition would vary depending on the initial influent PO4-P concentration.

CONCLUSIONS

A phosphorus crystallization-filtration (PCF) process was designed and operated as an alternative process for the removal of phosphorus from domestic wastewater as hydroxyapatite (HAP). During laboratory-scale column tests, Ca2+ dosage supplementation greatly enhanced phosphorus removal efficiency, and under a residual Ca2+ condition >80 mg/L, T-P and PO4-P, removal efficiencies above 95% were realized in several tests regardless of the pH condition or the linear velocity. Based on the results of the laboratory experiments, pre-treatment conditions (pH, residual Ca2+), filter depth, and linear velocity (EBCT and flow rate) are suggested as the major design and operating parameters of the PCF process. A successful pilot-scale operation showed that PCF is a promising process for phosphorus removal from domestic wastewater. After the PCF process, the average effluent concentrations of T-P, PO4-P, and SS were 0.05, 0.04, and 1.1 mg/L, respectively, corresponding to removal efficiencies of 90.0, 86.5, and 79.7%, respectively. Finally, the effluent T-P contents <0.2 mg/L and SS contents <10 mg/L obtained from the pilot-scale process satisfied Korean regulation regarding water resource conservation. Furthermore, the characteristics of the sludge obtained after the PCF process using influents with various initial PO4-P contents were investigated using SEM, FT-IR, EDS, and XRD analyses. Considering the high content of calcite and hydrated phosphorus compounds in the sludge recovered from the PCF process, the sludge can be used as an alternative soil amendment or fertilizer for the neutralization of acidic soils.

ACKNOWLEDGEMENTS

This research was funded by Korea Institute of Civil Engineering and Building Technology (KICT) under the grant number 20200039-001.

REFERENCES

REFERENCES
Bal Krishna
K. C.
Aryal
A.
Jansen
T.
2016
Comparative study of ground water treatment plants sludges to remove phosphorous from wastewater
.
Journal of Environmental Management
180
,
17
23
.
Berg
U.
Donnert
D.
Weidler
P. G.
Kaschka
E.
Knoll
G.
Nüesch
R.
2006
Phosphorus removal and recovery from wastewater by tobermorite-seeded crystallisation of calcium phosphate
.
Water Science and Technology
53
(
3
),
131
138
.
Chen
X.
Kong
H.
Wu
D.
Wang
X.
Lin
Y.
2009
Phosphate removal and recovery through crystallization of hydroxyapatite using xonotlite as seed crystal
.
Journal of Environmental Sciences
21
(
5
),
575
580
.
Cichy
B.
Kużdżał
E.
Krztoń
H.
2019
Phosphorus recovery from acidic wastewater by hydroxyapatite precipitation
.
Journal of Environmental Management
232
,
421
427
.
Deliyanni
E. A.
Peleka
E. N.
Lazaridis
N. K.
2007
Comparative study of phosphates removal from aqueous solutions by nanocrystalline akaganéite and hybrid surfactant-akaganéite
.
Separation and Purification Technology
52
(
3
),
478
486
.
Donnert
D.
Salecker
M.
1999
Elimination of phosphorus from waste water by crystallization
.
Environmental Technology
20
(
7
),
735
742
.
Egle
L.
Rechberger
H.
Zessner
M.
2015
Overview and description of technologies for recovering phosphorus from municipal wastewater
.
Resources, Conservation and Recycling
105
,
325
346
.
Guan
W.
Ji
F. Y.
Chen
Q. K.
Yan
P.
Zhou
W. W.
2014
Phosphorus recovery using porous calcium silicate hydrate as seed crystal in form of hydroxyapatite
.
Materials Research Innovations
18
(
1
),
43
49
.
Ishikawa
M.
Ichikuni
M.
1981
Coprecipitation of phosphate with calcite
.
Geochemical Journal
15
(
5
),
283
288
.
Kim
B. C.
Sa
S. H.
Kim
M. S.
Lee
Y. K.
Kim
J. K.
2007
The limiting nutrient of eutrophication in reservoirs of Korea and the suggestion of a reinforced phosphorus standard for sewage treatment effluent
.
Journal of Korean Society on Water Quality
23
(
4
),
512
517
.
Lee
K. W.
Lee
N. K.
Kim
Y. H.
2004
Assessment for management of village scale sewage plant in Chonnam Province
.
The Society For Environmental Technology In Korea
5
(
1
),
62
66
.
Menglin
Y.
Danyang
Y.
Jing
S.
Duanmei
S.
Zhengwen
X.
2016
Phosphorus removal and recovery from high phosphorus wastewater by the HAP crystallization process
.
Oriental Journal of Chemistry
32
(
1
),
235
241
.
Momberg
G. A.
Oellermann
R. A.
1992
The removal of phosphate by hydroxyapatite and struvite crystallization in South Africa
.
Water Science and Technology
26
(
5–6
),
987
996
.
Moriyama
K.
Kojima
T.
Minawa
Y.
Matsumoto
S.
Nakamachi
K.
2001
Development of artificial seed crystal for crystallization of calcium phosphate
.
Environmental Technology
22
(
11
),
1245
1252
.
Moriyama
K.
Kojima
T.
Koga
K.
Takino
S.
Minawa
Y.
2003
Crystallisation process using calcium silicate hydrate for phosphorus removal
.
Environmental Engineering Research
40
,
389
394
.
Nancollas
G. H.
1968
Kinetics of crystal growth from solution
.
Journal of Crystal Growth
3–4
,
335
339
.
Song
Y.
Hahn
H.
Hoffmann
E.
2001
The effect of carbonate on the precipitation of calcium phosphate
.
Environmental Technology
23
(
2
),
207
222
.
Song
Y.
Weidler
P. G.
Berg
U.
Nüesch
R.
Donnert
D.
2006
Calcite-seeded crystallization of calcium phosphate for phosphorus recovery
.
Chemosphere
63
(
2
),
236
243
.
Tozsin
G.
Arol
A. I.
Oztas
T.
Kalkan
E.
2014
Using marble wastes as a soil amendment for acidic soil neutralization
.
Journal of Environmental Management
133
(
15
),
374
377
.

Author notes

First author: Hyangyoun Chang