Science of the Total Environment 705 (2020) 135959

Contents lists available at ScienceDirect

Science of the Total Environment

j ourna l homepage: www.e lsev ie r .com/ locate /sc i totenv

Conversion of municipal wastewater-derived waste to an adsorbent forphosphorus recovery from secondary effluent

Wen-Jing Xia, Lian-Zeng-Ji Xu, Lin-Qian Yu, Quan Zhang, Yi-Heng Zhao, Jin-Rui Xiong, Xiao-Yan Zhu,Nian-Si Fan, Bao-Cheng Huang ⁎, Ren-Cun JinLaboratory of Environmental Technology, School of Life and Environmental Sciences, Hangzhou Normal University, Hangzhou 311121, China

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Iron enriched sludge was successfullyconverted into phosphorus absorbent.

• Phosphorus in secondary biological ef-fluent was readily reduced to b0.1 mg/L.

• The potential of reusing adsorbed phos-phorus as fertilizer was evaluated.

⁎ Corresponding author.E-mail address: huangbc@hznu.edu.cn (B.-C. Huang).

https://doi.org/10.1016/j.scitotenv.2019.1359590048-9697/© 2019 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 18 September 2019Received in revised form 16 November 2019Accepted 4 December 2019Available online 5 December 2019

Editor: Yifeng Zhang

Keywords:Alkali-treatmentIron sludgeMunicipal wastewaterPhosphate adsorptionSecondary effluent

The sustainable management and recirculation of phosphorus resources are essential to our human lives. In thiswork, phosphorus removal and recovery from secondary effluent were achieved using municipal wastewater-derived materials as adsorbents. Through modification with 0.5 M NaOH for 30 min, iron containing sludgethat originated from the coagulation pretreatment of municipal wastewater was successfully converted to phos-phorus adsorbent. The maximal adsorption capacity of the prepared adsorbent was estimated to be 22 mg-P/g,and the adsorption performance remained stable in the pH range of 5–8. FeO(OH) was identified as the key ad-sorption site, and the ligand exchange mediated chemical adsorption was the main mechanism for phosphorusremoval by the preparedmaterial. Moreover, a laboratory-scale continuous-flow adsorption column experimentshowed that the surplus phosphorus in secondary effluent could be readily reduced to b0.1 mg/L. By pyrolysis ofP-laden alkali-treated iron sludge under oxygen limited conditions, the phosphorus was recovered and success-fully applied to supportwheat growth. Thiswork provides valuable information for both the sustainablemanage-ment of phosphorus streams in wastewater and cyclic utilization of waste sludge.

© 2019 Elsevier B.V. All rights reserved.

1. Introduction

As an essential nutrient, phosphate (P) is vital to our human lives. Itwas estimated that people excreted approximately 98%of the P consumed

from food daily, accounting for approximately 3.3 Tg P per year in total(Yuan et al., 2018). Most of the excreted P finally ends up at wastewatertreatment plant (Cordell et al., 2009). Therefore, the removal of P frommunicipal wastewater is of great significance since excess phosphate insurface waters can result in eutrophication. Eutrophication of lakes andother water bodies has long been identified as a global environmentalissue and a great threat to aquatic ecosystems (Le Moal et al., 2019).

2 W.-J. Xia et al. / Science of the Total Environment 705 (2020) 135959

At present, biological phosphorus removal craft is still one of themostwidely implemented technologies in municipal wastewater treatmentplants (Li et al., 2019; Oehmen et al., 2007). Although biological phospho-rus removal processes are economically efficient, the treatment perfor-mance can be easily influenced by varied conditions. Effluentphosphorous concentration from conventional biological processesmight notmeet the discharge limit requirement, especially for thewaste-waters with low organic content. However, in countries or regions whereenvironmental quality is poor, strict control of phosphate emission is es-sential. For example, in China, conventional biological phosphorus re-moval technology could barely meet the national Class 1A dischargestandards (0.5 mg/L total P in total) (Sun et al., 2017). Nevertheless, sev-eral big cities, including Beijing and Tianjin, have already implementedmore stringent local total P emission standards than the current nationalone, inwhich the effluent total P b 0.3mg/L ismandatory. Forwastewatertreatment plants in these cities, cost-effective upgrades to reduce the bio-logical effluent phosphate concentration to meet the local emission re-quirements were substantially challenged (Li et al., 2012).

For low phosphorus content in the secondary effluent, adsorption isone of the most efficient and economic removal and recovery methods(Huang et al., 2019). The synthesis and preparation of phosphorus adsor-bents with high efficiency and good regeneration ability are substantialfor practical applications of this process. Previous studies have widelydocumented that metal salt (iron, aluminium, magnesium, lanthanum,etc.) embodied adsorbents were feasible for phosphorus removal(Mayer et al., 2013; Hao et al., 2019; Zhang et al., 2012). To improvema-terial recovery, composite metal adsorbents with magnetic propertieshave been designed, and application feasibility has been tested in apilot-scale experiment (Drenkova-Tuhtan et al., 2017; Rott et al., 2018).In addition to regeneration, the initial investment of rawmaterial for ad-sorbent fabrication is another factor that could affect the economics.Reusing waste residues as raw constituents to prepare phosphorus ad-sorbent might provide a solution to address the above concern.

To enhance the effectiveness of biological phosphorus removal pro-cesses, iron-based coagulant has beenwidely applied inmunicipalwaste-water treatment plants, either before or after the biological treatment unit(Huang et al., 2017; Wang et al., 2014). In rapidly growing cities, chemi-cally enhanced primary treatment is used to release the plant's burden(Lin et al., 2017; Wang et al., 2009). Furthermore, for sustainable treat-ment of municipal wastewater, iron-based coagulants are thought to bemore popular for use in upcoming wastewater treatment plants(Wilfert et al., 2015). Thus, wide application of coagulants will inevitablylead to production of large amounts of iron-containing sludge. Appropri-ate treatment and disposal of inorganic component mixed sludge wouldbe essential. Iron-enriched sludge derived fromthemunicipalwastewatermight be appropriately reused as adsorbent for the removal of low con-centration P from secondary effluent. Through further appropriate treat-ment, phosphorus recovery can be expected to mitigate phosphorusshortages due to its over-exploitation (Liu et al., 2016;Mayer et al., 2016).

Based on the above considerations, the main purpose of this work isto evaluate the feasibility and performance of reusingmunicipal waste-water derived iron-containing sludge (IS) as phosphorus adsorbent. Al-kali pretreatment was employed as the modification method, and thesurface characteristics and adsorption performance of alkali treated IS(AIS) was explored. Additionally, possible adsorption mechanismswere investigated. Lastly, the feasibility of utilizing P-laden AIS (PAIS)as a potential phosphate fertilizer was discussed. This work providesvaluable information for metal-containing wastes disposition and sus-tainable management of phosphorus streams in wastewater.

2. Materials and methods

2.1. AIS preparation

The raw influent from a local Qige sewage treatment plant (Hang-zhou, China) was collected and used to produce IS. In brief, FeCl3·6H2O

with a final content of 30 mg/L Fe was added to the wastewater to startthe coagulation reaction. After 300 rpm of rapid agitation for 5 min and60 rpm of slow agitation for 25 min, precipitates were collected anddried for 12 h at 50 °C to obtain IS. The chemical composition of thewastewater is shown in Table S1.

In this work, alkali treatment was adopted as the modificationmethod to improve the adsorption capacity of IS. Briefly, the dewateredIS powder was soaked in NaOH solution under magnetic stirring condi-tion for a suitable time. Afterwards, the materials were collected andrinsed with deionized water until the pH of the leaching solution wasneutral. After dehydrating the abovematerial at 50 °C, AISwas obtained.Since the concentration of the alkali solution and contact times mightgreatly influence the adsorption ability, these two preparation parame-ters were optimized. Specifically, the effects of three concentrations ofalkali solution, i.e., 0.5, 1, and 2 M NaOH, and five contact times,i.e., 10, 30, 60, 120, and 360 min, on the product's adsorption perfor-mance were investigated.

2.2. Batch and continuous-flow experiments for phosphorus adsorption

In this work, batch adsorption experiments were initially carried outin an oscillator at 30± 1 °C and 180 rpm to evaluate the adsorption per-formance of the as-preparedmaterial. In the typical adsorption process,25 mg adsorbent was dosed into 50 mL phosphate solution (20 mg/Lphosphorus in the form of KH2PO4). The aqueous pH was adjusted to7.0 ± 0.2 by adding 1 M HCl or NaOH. The adsorption properties ofAIS under other pH conditions (i.e., 5.0, 6.0, 7.0 and 8.0) were alsotested. In addition, different anions (0.1 M Cl−, NO3

−, SO42−, and CO3

2−)were added to thephosphorus stock solution to explore the interferenceof these ions on phosphate removal. All experiments were conductedrepeatedly, and the averaged results are reported in the present work.

To test the feasibility of using AIS as an adsorbent for treatingphosphorus-containing wastewater, continuous-flow adsorption ex-periments in a fixed-bed column (Φ 1 × 10 cm) were conducted. Theschematic diagram of the adsorption column is shown in Fig. S1. Theas-prepared AIS was added to the column with a 10% volume fillingratio. Biological secondary effluent collected from the Qige wastewatertreatment plant with approximately 0.2 mg/L phosphate was pumpedupward through the adsorption column (0.25mL/min) at room temper-ature. The Thomas model (Eq. (1)) (Malkoc et al., 2006) was employedto analyse the column experimental results.

C0

Ct¼ exp

Kthv

mQe−C0VEff� �� �

þ 1 ð1Þ

where Kth (mL/min/mg) represents the Thomas rate constant, VEff (L) isthe effluent volume, v (mL/min) is flow rate, V (cm3) is the volume ofadsorbent bed, Qe (mg/g) is the amount of equilibrium phosphate ad-sorption, and m (g) is the adsorbent mass.

2.3. Adsorption models

Adsorption kinetics were fitted by the pseudo-first-order (Eq. (2))(Ho and McKay, 1998) and pseudo-second-order equations (Eq. (3))(Ho and McKay, 1999). The Weber and Morris intraparticle diffusionmodel (Eq. 4) (Hamayun et al., 2014) was also used to investigatewhether the intraparticle diffusion existed during the adsorption pro-cess.

log Q–Qtð Þ ¼ logQe–kt=2:303 ð2Þ

tQt

¼ 1k2Qe

þ 1Qe

t ð3Þ

Qt ¼ kidt1=2þ C ð4Þ

3W.-J. Xia et al. / Science of the Total Environment 705 (2020) 135959

where Qt (mg/g) is the adsorbed phosphate amount at reaction time t, kis the kinetic constant, and C is the constant.

Freundlich (Eq. (5)) and Langmuir adsorption isotherms (Eq. (6))(Tahir and Rauf, 2006; Yoon et al., 2016) were applied to analysis thephosphate adsorption by AIS.

logQe ¼ logK f þ1n

logCe ð5Þ

Ce

Qe¼ 1

AQmaxþ Ce

Qmaxð6Þ

where Ce is the phosphate content (mg/L) when adsorption reachesequilibrium, Qmax is the maximal phosphorous adsorption capacity(mg/g), A is the adsorption free energy (L/mg), Kf can express the ad-sorption capacity roughly, and n represents an empirical parameterthat indicates the dependence of adsorption on equilibriumconcentration.

Adsorption thermodynamics were studied by exploring the phos-phorus removal performance at different temperatures (293, 303, 313,and 333 K) with an original 20 mg-P/L phosphate concentration and0.5 g/L AIS. The activation energy (Ea) was estimated using the obtainedrate constants under different temperature conditions (Eq. (7)). Otherthermodynamic indexes were counted following Eqs. (7)–(9). (Al-Ghouti et al., 2005; Mahmood et al., 2011):

lnk ¼ lnK0 þ EaRT

ð7Þ

logQe ¼ΔS

2:303Rþ ΔH2:303RT

ð8Þ

ΔG ¼ ΔH−TΔS ð9Þ

where K0 is the frequency factor, R is the ideal gas constant, and T is the

Fig. 1. Phosphorus adsorption characteristics of the prepared adsorbent. (a) Adsorption kineticadsorption by AIS, and (d) the Arrhenius plot of lnk versus 1/T to determine activation entemperature for (a), 293, 303, 318, and 333 K temperature for (b), and initial pH = 7.0.

Kelvin temperature. ΔG, ΔH, and ΔS are the changes in Gibbs free en-ergy, entropy and enthalpy, respectively, during the adsorption process.

2.4. Wheat seedling cultivation

To explore whether adsorbed phosphate could be reused as a poten-tial fertilizer, wheat cultivation experiments were conducted. First, thePAIS acquired from the continuous-flow column was carbonated at400 °C under N2 atmosphere for 2 h to prepare a phosphorus-rich bio-char. Then, the biochar was used as an alternative nutrient source forwheat growth. In this work, to exclude the interferences of residual or-ganic and inorganic constituents on wheat growth, glass sand was se-lected as the simulated soil. Wheat cultivation experiments consistedof one experimental group (with biochar addition) and two controlgroups (without biochar addition). For the experimental group, 10%phosphorus-rich biochar was applied as a phosphorus alternative.Each group had two replicateswith 15 germinatedwheat seeds planted.Modified Hoagland nutrient solutionwas used in this work. For the bio-char dosed group, P-deficient modified Hoagland solution was applied,and it was denoted as biochar-Plack. The experimental group with themodified Hoagland solution but no biochar was denoted as Glasssand-Pfull. In comparison, the last group with P-deficient modifiedHoagland solution and no biochar was denoted as Glass sand-Plack. De-tailed information of the experimental design can be found inTable S2. All the cultivation conditions (temperature, humidity and sun-light) remained the same for the three groups. After 10 days, 13 seed-lings were randomly selected for growth status analysis, includingplant height, root length and raw weight. In addition, the content ofcommon heavy metals in P-rich biochar was determined.

2.5. Characterization and analytical methods

The scanning electron microscopy images (SEM, Nova 450, FEI Co.,USA) observed the surface morphology and microscopic imaging ofthe adsorbent. In addition, an energy dispersive spectrometer (EDS,

s of AIS and IS, (b) the Weber-Morris model, (c) effects of temperature on the phosphateergy. Experimental conditions: 20 mg/L initial phosphorus, 0.5 g/L adsorbent, 303 K

Table 2Thermodynamic data for the adsorption of phosphate onto AIS.

C0(mg/l)

Ea(kJ/mol)

ΔS(J/mol K)

ΔH(KJ/mol)

ΔG (J/mol K)

293 K 303 K 318 K 333 K

20.00 26.00 78.85 26.07 10.12 6.86 3.87 −0.95

4 W.-J. Xia et al. / Science of the Total Environment 705 (2020) 135959

INCA, England) was used to analyse the element types and contents inmicroregions. The characteristics and structural information of surfacefunctional groupswere analysed byNicoletis 10 Fourier transform infra-red spectroscopy (FT-IR, Thermo Fisher Nicolet, USA). X-ray diffraction(XRD, D8 Advance, Germany) was applied to acquire the materialphase. X-ray photoelectron spectroscopy (XPS, ESCALAB, ThermoFisher, Inc., USA) determined the chemical valence and characteristicsof iron, phosphorus and oxygen on the adsorbent surface before andafter adsorption.

All samples were withdrawn and filtered with a polytetrafluoroethylenemembrane (0.45 μm) before analysis. Phosphorus content was mea-sured using the molybdenum antimony anticolorimetric method(APHA et al., 2005), andmetal elementswere analysed by an atomic ab-sorption spectrophotometer (AAS, AA6800, Japan).

3. Results and discussion

3.1. Phosphorus adsorption properties of prepared adsorbents

3.1.1. Adsorption kinetics and thermodynamicsBefore evaluating the adsorption kinetics of AIS, the preparation

condition (i.e., alkali concentration and treatment time) was optimized.As shown in Fig. S2, 30-min contact time and 0.5 M NaOH wasascertained as the optimal conditions for AIS preparation. Typical phos-phorus adsorption curves of the prepared AIS and IS are shown inFig. 1a. Nearly 90% of the Qe was completed during the initial 2 h, andit took 3.5 h to reach adsorption equilibrium for AIS. As a comparison,IS required 200 min to reach 90% adsorption capacity and 5 h to reachadsorption equilibrium. By fitting the experimental data to the kineticdesorptionmodels, the results of the phosphorus adsorption kinetics in-dicated that therewas a low correlation between the pseudo-first-orderkineticmodel and the data, as evidenced by the poor R2 values (R2 b 0.8)for both AIS and IS. In comparison, the pseudo-second order modeldemonstrated a good fit to the adsorption kinetics data (R2 N 0.9). Thevalues of other important kinetic parameters are listed in Table 1. Sim-ilar to a result reported by Rashid et al. (2017), rising temperatures ac-companied increasing rate constant (k2) values. Additionally, highertemperatures resulted in a higher adsorption capacity, and these resultsshowed a typical chemisorption process where electron sharing or ex-change occurred between phosphate and functional groups of AIS(Chiew et al., 2009; Gücek et al., 2005). A well-fitting result by thepseudo-second order model indicated that phosphorus was removedby AIS mainly by the chemisorption process (Ling et al., 2017).

The adsorption process was further discussed, and the kinetic datawere tested by theWeber andMorris model. In this study, a high corre-lation (R2 N 0.9) existed between the experimental data and theWeber-Morris model (Fig. 1b), indicating that intraparticle diffusion played arole during the AIS phosphate adsorption processes. However, thefitting line did not pass through the origin of the axes, indicating that in-terparticle diffusion was not a rate-limiting step controlling the adsorp-tion process (Al-Duri and McKay, 1988; Yang et al., 2018).

By employing the Arrhenius equation to simulate the fitted rate con-stants obtained under different temperature conditions (Fig. 1c), a serialof thermodynamic parameters were acquired (Table 2). The plot inFig. 1d shows a well-fitting result of the Arrhenius equation, and the

Table 1Adsorption kinetic parameters of AIS with different temperatures.

InitialphosphateConc.(mg/L)

Qe

(mg/g)Temperature(K)

Pseudo-first orderkinetics

R2 Qe-1 K1 R2 Qe-2 K2

20 10.4 293 0.886 8.55 0.15 0.916 11.05 0.0113.9 303 0.947 9.28 0.02 0.974 13.86 0.0217.5 318 0.959 10.23 0.09 0.995 18.04 0.0324.0 333 0.960 13.55 0.08 0.986 23.20 0.04

Ea of phosphorus adsorption by AIS was calculated to be 26 kJ/mol.These results again indicated that phosphorus adsorption by AIS wasdriven by chemisorption (Al-Ghouti et al., 2005; Liu et al., 2011). Thepositive value of ΔH indicated that phosphorus removal was an endo-thermic process. High temperaturesmay have been beneficial to the re-action between phosphate and AIS owing to the reactive energy barrierreduction and thereby favoured the mass transport. A positive ΔS im-plied that the phenomenon of chaos on the solid-solution interface in-creased, suggesting the adsorption of phosphate on the surface of AISwas promoted (Lazaridis et al., 2003; Rashid et al., 2017). Furthermore,with low temperature, the positive ΔG indicated an energy barrierexisted when the chemical reaction would take place. Moreover, thevalues of ΔG decreased with increasing temperature. Additionally, ahigh temperature could accelerate the diffusion of phosphate into theAIS interior by providing more energy and then promoting the adsorp-tion rate (Mahmood et al., 2011).When the temperature reached 333K,the value ofΔG becamenegative,which implied that adsorption processhad become a spontaneous process (Rashid et al., 2017).

3.1.2. Adsorption isothermWith 0.5 g/L AIS, the initial phosphate concentration was increased

from 2 mg/L to 50 mg/L to investigate the AIS adsorption isotherm. Asshown in Table 3, the Freundlich isotherm model was more suitablefor explaining the adsorption process than that of Langmuir, as evi-denced by its higher R2 value (R2 N 0.94). Hence, the above results im-plied that phosphorus adsorption by AIS was a multilayer adsorptiondominated process. Low values of 1/n (b0.40) implied that chemisorp-tion was involved during the phosphorus adsorption process (Yanget al., 2018).

3.2. Adsorption mechanism

Themicrostructure of IS and AISwere characterized by SEM imaging(Fig. 2a and b). AIS had a rougher and more porous surface than IS. Thisresult indicated that alkali treatmentwas beneficial to porosity develop-ment. The porous structure was thought to improve the surface and fa-cilitate the phosphate adsorption (Loganathan et al., 2014; Zhang et al.,2012). EDS analysis further revealed that the adsorbent was mainlycomposed of C, O and Fe elements (Fig. 2c and d).

After AIS adsorbed phosphate at pH 7.0, the FTIR spectra appeared asa stronger peak at 1033 cm−1 corresponding to the band vibration ofHPO4

2−or H2PO4− (Elzinga and Sparks, 2007), which was expected with

the adsorbed phosphate species. The XRD patterns of AIS and PAISwith a 2θ range between 10° to 80° are illustrated in the Fig. 3b. Thepeaks at 2θ values of 26.7° were attributed to the reflection of FeO(OH) (Yang et al., 2018). XPSwas further used to characterize the adsor-bent structure variation before and after the phosphorus adsorption re-action (Fig. 3c and d). High-resolution deconvoluted O1s spectra showedthat elemental O within the adsorbent was mainly in the forms O2–

(530.3 eV), -OH (531.7 eV), and C\\O (533.1 eV) (Fang et al., 2017;

Table 3Langmuir and Freundlich parameters for the adsorption of phosphate by AIS.

Langmuir isotherm Freundlich isotherm

R2 qmax Ka R2 Kf 1/n

0.79 22.17 0.13 0.94 2.29 0.38

Fig. 3. Chemical structure and composition characterizations. (a) FT-IR, (b) XRD, and High-resolution deconvoluted O1s spectra of (c) AIS and (d) PAIS.

Fig. 2. SEM–EDS characterization of (a, c) IS and (b, d) AIS.

5W.-J. Xia et al. / Science of the Total Environment 705 (2020) 135959

Table 4Growth attributes of wheat seedlings.

Groups Plant height (cm) Root length (cm) Wet weight (g)

Glass sand-Pfull 12.5 ± 0.3 6.4 ± 0.5 0.20 ± 0.05Biochar-Plack 14.7 ± 0.2 6.5 ± 0.4 0.22 ± 0.02Glass sand-Plack 7.1 ± 0.3 3.4 ± 0.4 0.11 ± 0.04

Fig. 4. (a) Effect of initial pH on the adsorption of phosphate onto AIS and IS, (b) effect ofcoexisting ions for AIS phosphate adsorption, and (c) reuse stability of AIS. Experimentalconditions: 0.5 g/L adsorbent, 303 K temperature, simulated wastewater with 20 mg/Linitial phosphorus and pH= 7.0 for (a) and (b) and practical secondary effluent for (c).

6 W.-J. Xia et al. / Science of the Total Environment 705 (2020) 135959

Qiu et al., 2015). After phosphorus was adsorbed onto the material sur-face, the proportion of -OHwas decreased from 81 at.% to 56 at.%. In ac-cordance, O2– content was increased from 13 at.% to 34 at.%. The aboveresults indicated that -OH was involved in phosphorus adsorption,and replacement of -OH by PO4

3− resulted in its reduction.

Fig. 5. Breakthrough curve of phosphate adsorption on AIS. Experimental conditions: 10%volume filling ratio of adsorbent, room temperature, and HRT = 40 min.

According to the comprehensive FTIR and XRD analyses, iron wasmainly existed in the form of FeO(OH). XPS characterization furtherdemonstrated that hydroxyl participated in the adsorption reactions.Thus, it could be concluded that ligand exchangemediated chemical ad-sorption was the main mechanism for phosphorus removal by AIS(Zhang et al., 2009). By combining Ea and the Freundlich isothermmodel, the exchange of ligands between OH– and phosphate anionswas proposed as follows (Vikrant et al., 2018):

AIS-3FeO(OH) + PO43− → AIS-(FeO)3PO4 + 3OH− (10).

3.1. Removal and recovery of surplus phosphorus from secondaryeffluent

Before evaluating the feasibility of applying AIS to treat practical sec-ondary effluent, the influences of key factors (i.e., pH and anions) on theadsorption performance were explored. For the 5.0–8.0 pH range, AISshowed a stable phosphorus removal performance (Fig. 4a). Anions,such as sulfate, nitrate, chloride and carbonate, usually coexisted withphosphate (Namasivayam and Sangeetha, 2006; Yang et al., 2018),and their interferences on phosphorus adsorption were investigated.Monovalent anions were found to have a negligible impact on the ad-sorbent performance (Fig. 4b). In comparison, divalent anions couldgreatly inhibit the phosphorus removal by the prepared AIS. Similarphenomena were observed in other studies (Loganathan et al., 2014;Yang et al., 2018). Themain reasonwas that CO3

2– or SO42− could strongly

compete with phosphate for the adsorption sites on AIS.The effectiveness of the prepared adsorbent on phosphorus removal

was further examined using a local secondary effluent as an example.Since the reuse stability was vital for practical applications, the adsorp-tion performance of the adsorbent after differing numbers of use wasexplored. The results showed that although the adsorption performanceshowed a decline, N80% phosphorus removal efficiency was stillachieved after its third-cycle use (Fig. 4c).

By packing the prepared AIS into a glass column, the performance oftreating secondary effluentwas then evaluated in a continuous-flow re-actor. Fixed-bed column experimental results showed that the effluentphosphorus concentration could be readily controlled within 0.1 mg/L,even though it has been treated by 400 times the volume of the column(Fig. 5). The phosphorus concentration in the column effluent could re-liably meet the Class II quality of China Surface Water EnvironmentalQuality standards. The maximum capacity of the adsorption columnwas estimated to be 20.56 mg/g according to the Thomas model (Ta-ble S3), which was slightly lower than that in the batch experiment(22.17 mg/g). The above results indicated that the prepared adsorbentwas efficient for removing surplus phosphorus in secondary effluent.

To further explore the feasibility of reusing adsorbed phosphorus asa potential fertilizer, the wheat seedling growth experiment was con-ducted, and the results are shown in Table 4. After 10-days of growth,the height, length, and wet weight of wheat dosed with 10% P contain-ing biocharwere 14.7± 0.2 cm, 6.5±0.4 cm, and 0.22±0.02 g, respec-tively. The growth status of wheat dosed with biochar was comparableto that of the group fertilized with phosphorus solution. However, incomparison, when phosphorus was limited, a retarded growth was ob-served. Thus, the above results indicated that recovered phosphorusfrom secondary effluent could be used as an alternative phosphate soilconditioner to support plant growth.

7W.-J. Xia et al. / Science of the Total Environment 705 (2020) 135959

3.3. Implications

Phosphorus is a key element for inducing the eutrophication of sur-face water. Additionally, it is a non-renewable resource for supportingfood production. Therefore, sustainable management of daily producedwaste phosphorus is meaningful. The present work demonstrated thatphosphorus can be readily removed and recovered as an alternative fer-tilizer from secondary effluent using a municipal wastewater derivedadsorbent. Therefore, not only was wasted sludge reused, but circula-tion of surplus phosphoruswas also achieved. Direct application of sew-age sludge as fertilizer on agriculture field is the commonlyimplemented practice in the Europe (Fytili and Zabaniotou, 2008;Hudcova et al., 2019). However, the risks of potential toxic metalsleaching and pathogen challenge its application. In comparison, pyroly-sis treatment of sludge favours for the toxic organic pollutants removaland reduced metal availability (Mendez et al., 2012; Hoffman et al.,2016). Song et al. (2014) observed that heavy metal accumulated inthe plantwas greatly inhibited via pyrolysis of sludge under appropriatetemperature. Moreover, it has been revealed that pyrolysis treatment isbeneficial to increase the bio-availability of phosphorus (Fristak et al.,2018).

In China, for example, approximately 1378 t total phosphorus wasdischarged with the secondary effluent nationally in 2018 (http://www.mohurd.gov.cn/csjs/xmzb/index.html). If the proposed recoveryroute was applied to control effluent phosphorus to b0.1 mg/L, a totalof 951 t phosphorus could be efficiently recovered. Heavy metal con-tents within the biochar might be of potential concern for its applica-tion. However, in the present work, the heavy metal contents inpyrolysis biochar (shown in Table S4) were below the upper limitingvalue according to the International Biochar Initiative report (IBI, 2015).

4. Conclusions

In thiswork,municipalwastewater derived iron sludgewas success-fully converted into phosphorus adsorbent through alkali modification.The maximal adsorption capacity of the prepare adsorbent was esti-mated to reach 22 mg-P/g, and the adsorption performance remainedstable in the pH range of 5 to 8. The Freundlich model isothermwas ap-plied to describe the phosphorus adsorption on adsorbent surface. Bycombing different characterizationmethods, ligand exchangemediatedchemical adsorption was the main mechanism for phosphorus removalby the prepared material. Moreover, FeO(OH) was identified as the keyadsorption site. Laboratory-scale continuous-flow experiments indi-cated that surplus phosphorus in the secondary effluent could be readilyreduced to b0.1 mg/L, which meets the Class II quality of China SurfaceWater Environmental Quality standards. By pyrolysis of P-laden alkali-treated iron sludge under oxygen limited condition, P embodied biocharwas successfully obtained and further reused as a potential phosphatesoil conditioner to support plants growth.

Declaration of competing interest

The authors declare that they have no known competing financialinterests or personal relationships that could have appeared to influ-ence the work reported in this paper.

Acknowledgments

The authors thank the Natural Science Foundation of Zhejiang Prov-ince (LQ19E080015) for supporting this work.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.scitotenv.2019.135959.

References

Al-Duri, B., McKay, G., 1988. Basic dye adsorption on carbon using a solid-phase diffusionmodel. Chem. Eng. J. 38, 23–31.

Al-Ghouti, M., Khraisheh, M.A., Ahmad, M.N., Allen, S., 2005. Thermodynamic behaviourand the effect of temperature on the removal of dyes from aqueous solution usingmodified diatomite: a kinetic study. J. Colloid Interface Sci. 287, 6–13.

APHA, AWWA, AEF, 2005. StandardMethods for the Examination ofWater andWastewa-ter. American Public Health Association, Washington, DC, USA.

Chiew, H., Sampson, M.L., Huch, S., Ken, S., Bostick, B.C., 2009. Effect of groundwater ironand phosphate on the effificacy of arsenic removal by iron-amended biosand fifilters.Environ. Sci. Technol. 43, 6295–6300.

Cordell, D., Drangert, J.O., White, S., 2009. The story of phosphorus: global food securityand food for thought. Glob. Environ. Chang. 19, 292–305.

Drenkova-Tuhtan, A., Schneider, M., Franzreb, M., Meyer, C., Gellermann, C., Sextl, G.,Mandel, K., Steinmetz, H., 2017. Pilot-scale removal and recovery of dissolved phos-phate from secondary wastewater effluents with reusable ZnFeZr adsorbent @Fe3O4/SiO2 particles with magnetic harvesting. Water Res. 109, 77–87.

Elzinga, E.J., Sparks, D.L., 2007. Phosphate adsorption onto hematite: an in situ ATR-FTIRinvestigation of the effects of pH and loading level on the mode of phosphate surfacecomplexation. J. Colloid Interface Sci. 308, 53–70.

Fang, L.P., Wu, B., Lo, M.C., 2017. Fabrication of silica-free superparamagnetic ZrO2@Fe3O4

with enhanced phosphate recovery from sewage: performance and adsorptionmechanism. Chem. Eng. J. 319, 258–267.

Fristak, V., Pipiska, M., Soja, G., 2018. Pyrolysis treatment of sewage sludge: a promisingway to produce phosphorus fertilizer. J. Clean. Prod. 172, 1772–1778.

Fytili, A., Zabaniotou, A., 2008. Utilization of sewage sludge in EU application of old andnew methods-a review. Renew. Sust. Energ. Rev. 12, 116–140.

Gücek, A., Şener, S., Bilgen, S., Mazmanci,M.A., 2005. Adsorption and kinetic studies of cat-ionic and anionic dyes on pyrophyllite from aqueous solutions. J. Colloid Interface Sci.286, 53–60.

Hamayun, M., Mahmood, T., Naeem, A., Muska, M., Din, S.U., Waseem, M., 2014. Equilib-rium and kinetics studies of arsenate adsorption by FePO4. Chemosphere 99,207–215.

Hao, H., Wang, Y., Shi, B., 2019. NaLa(CO3)2 hybridized with Fe3O4 for efficient phosphateremoval: synthesis and adsorption mechanistic study. Water Res. 155, 1–11.

Ho, Y.S., McKay, G., 1998. A comparison of chemisorption kinetic models applied to pol-lutant removal on various sorbents. Process. Saf. Environ. Prot. 76, 332–340.

Ho, Y.S., McKay, G., 1999. Pseudo-second order model for sorption processes. ProcessBiochem. 34, 451–465.

Hoffman, T.C., Zitomer, D.H., McNamara, P.J., 2016. Pyrolysis of wastewater biosolids sig-nificantly reduces estrogenicity. J. Hazard. Mater. 317, 579–584.

Huang, B.C., Guan, Y.F., Chen, W., Yu, H.Q., 2017. Membrane fouling characteristics andmitigation in a coagulation-assistedmicrofiltration process for municipal wastewaterpretreatment. Water Res. 123, 216–223.

Huang, X., Wu, W., Xia, Y., Li, W., Gong, Y., Li, Z., 2019. Alkali resistant nanocomposite gelbeads as renewable adsorbents for water phosphate recovery. Sci. Total Environ. 685,10–18.

Hudcova, H., Vymazal, J., Rozkosny, M., 2019. Present restrictions of sewage sludge appli-cation in agriculture within the European Union. Soil Water Res. 14, 104–120.

International Biochar Initiative (IBI), 2015. Standardized Product Definition and ProductTesting Guidelines for Biochar that Is Used in Soil (version number 2.1).

Lazaridis, N.K., Karapantsios, T.D., Georgantas, D., 2003. Kinetics of sorptive removal ofchromium(VI) from aqueous solutions by calcined Mg–Al–CO3 hydrotalcite. WaterRes. 37, 2875–2882.

Le Moal, M., Gascuel-Odoux, C., Menesguen, A., Souchon, Y., Etrillard, C., Levain, A.,Moatar, F., Pannard, A., Souchu, P., Lefebvre, A., Pinay, G., 2019. Eutrophication: anew wine in an old bottle? Sci. Total Environ. 615, 1–11.

Li, W.W., Sheng, G.P., Zeng, R.J., Liu, X.W., Yu, H.Q., 2012. China's wastewater dischargestandards in urbanization: evolution, challenges and implications. Environ. Sci. Pollut.Res. Int. 19, 1422–1431.

Li, Y.Y., Rahman, S.M., Li, G.Y., Fowle, W., Nielsen, P.H., Gu, A.Z., 2019. The composition andimplications of polyphosphate-metal in enhanced biological phosphorus removalsystems. Environ. Sci. Technol. 53, 1536–1544.

Lin, L., Li, R.H., Yang, Z.Y., Li, X.Y., 2017. Effect of coagulant on acidogenic fermentation ofsludge from enhanced primary sedimentation for resource recovery: comparison be-tween FeCl3 and PACl. Chem. Eng. J. 325, 681–689.

Ling, L.L., Liu, W.J., Zhang, S., Jiang, H., 2017. Magnesium oxide embedded nitrogen self-doped biochar composites: fast and high-efficiency adsorption of heavy metals inan aqueous solution. Environ. Sci. Technol. 51, 10081–10089.

Liu, J., Wan, L., Zhang, L., Zhou, Q., 2011. Effect of pH, ionic strength, and temperature onthe phosphate adsorption onto lanthanum-doped activated carbon fiber. J. Colloid In-terface Sci. 364, 490–496.

Liu, X., Sheng, H., Jiang, S., Yuan, Z., Zhang, C., Elser, J.J., 2016. Intensification of phosphoruscycling in China since the 1600s. Proc. Natl. Acad. Sci. 113, 2609–2614.

Loganathan, P., Vigneswaran, S., Kandasamy, J., Bolan, N.S., 2014. Removal and recovery ofphosphate from water using sorption. Crit. Rev. Environ. Sci. Technol. 44, 847–907.

Mahmood, T., Saddique, M.T., Naeem, A., Mustafa, S., Zeb, N., Shah, K.H., Waseem, M.,2011. Kinetic and thermodynamic study of cd(II), co(II) and Zn(II) adsorption fromaqueous solution by NiO. Chem. Eng. J. 171, 935–940.

Malkoc, E., Nuhoglu, Y., Abali, Y., 2006. Cr(VI) adsorption by waste acorn of Quercusithaburensis in fixed beds: prediction of breakthrough curves. Chem. Eng. J. 119,61–68.

Mayer, B.K., Gerrity, D., Rittmann, B.E., Reisinger, D., Brandt-Williams, S., 2013. Innovativestrategies to achieve low total phosphorus concentrations in high water flows. Crit.Rev. Environ. Sci. Technol. 43 (4), 409–441.

8 W.-J. Xia et al. / Science of the Total Environment 705 (2020) 135959

Mayer, B.K., Baker, L.A., Boyer, T.H., Drechsel, P., Gifford, M., Hanjra, M.A., Parameswaran,P., Stoltzfus, J., Westerhoff, P., Rittmann, B.E., 2016. Total value of phosphorus recov-ery. Environ. Sci. Technol. 50, 6606–6620.

Mendez, A., Gomez, A., Paz-Ferreiro, J., Gasco, G., 2012. Effects of sewage sludge biocharon plant metal availability after application to a Mediterranean soil. Chemosphere89, 1354–1359.

Namasivayam, C., Sangeetha, D., 2006. Removal of molybdate from water by adsorptiononto ZnCl2 activated coir pith carbon. Bioresour. Technol. 97, 1194–1200.

Oehmen, A., Lemos, P.C., Carvalho, G., Yuan, Z., Keller, J., Blackall, L.L., Reis, M.A., 2007. Ad-vances in enhanced biological phosphorus removal: from micro to macro scale.Water Res. 41, 2271–2300.

Qiu, H., Liang, C., Zhang, X., Chen, M., Zhao, Y., Tao, T., Xu, Z., Liu, G., 2015. Fabrication of abiomass-based hydrous zirconium oxide nanocomposite for preferable phosphate re-moval and recovery. ACS Appl. Mater. Interfaces 7, 20835–20844.

Rashid, M., Price, N.T., Gracia Pinilla, M., O’Shea, K.E., 2017. Effective removal of phosphatefrom aqueous solution using humic acid coated magnetite nanoparticles. Water Res.123, 353–360.

Rott, E., Nouri, M., Meyer, C., Minke, R., Schneider, M., Mandel, K., Drenkova-Tuhtan, A.,2018. Removal of phosphonates from synthetic and industrial wastewater with reus-able magnetic adsorbent particles. Water Res. 145, 608–617.

Song, X.D., Xue, X.Y., Chen, D.Z., He, P.J., Dai, X.H., 2014. Application of biochar from sew-age sludge to plant cultivation: Influence of pyrolysis temperature and biochar-to-soil ratio on yield and heavy metal accumulation. Chemosphere 109, 213–220.

Sun, J., Yang, Q., Wang, D., Wang, S., Chen, F., Zhong, Y., Yi, K., Yao, F., Jiang, C., Li, S., Li, X.,Zeng, G., 2017. Nickel toxicity to the performance and microbial community of en-hanced biological phosphorus removal system. Chem. Eng. J. 313, 415–423.

Tahir, S.S., Rauf, N., 2006. Removal of a cationic dye from aqueous solutions by adsorptiononto bentonite clay. Chemosphere 63 (11), 1842–1848.

Vikrant, K., Kim, K.H., Ok, Y.S., Tsang, D.C.W., Tsang, Y.F., Giri, B.S., Singh, R.S., 2018.Engineered/designer biochar for the removal of phosphate in water and wastewater.Sci. Total Environ. 616, 1242–1260.

Wang, H.T., Li, F.T., Keller, A.A., Xu, R., 2009. Chemically enhanced primary treatment(CEPT) for removal of carbon and nutrients from municipal wastewater treatmentplants: a case study of Shanghai. Water Sci. Technol. 60 (7), 1803–1809.

Wang, Y., Tng, K.H., Wu, H., Leslie, G., Waite, T.D., 2014. Removal of phosphorus fromwastewaters using ferrous salts-a pilot scale membrane bioreactor study. WaterRes. 57, 140–150.

Wilfert, P., Kumar, P.S., Korving, L., Witkamp, G.J., Loosdrecht, V.M.C., 2015. The relevanceof phosphorus and iron chemistry to the recovery of phosphorus from wastewater: areview. Environ. Sci. Technol. 49 (16), 9400–9414.

Yang, Q., Wang, X., Luo, W., Sun, J., Xu, Q., Chen, F., Zhao, J., Wang, S., Yao, F., Wang, D., Li,X., Zeng, G., 2018. Effectiveness and mechanisms of phosphate adsorption on iron-modified biochars derived from waste activated sludge. Bioresour. Technol. 247,537–544.

Yoon, Y., Park, W.K., Hwang, T.M., Yoon, D.H., Yang, W.S., Kang, J.W., 2016. Comparativeevaluation of magnetite-graphene oxide and magnetite-reduced graphene oxidecomposite for As(III) and As(V) removal. J. Hazard. Mater. 304, 196–204.

Yuan, Z., Jiang, S., Sheng, H., Liu, X., Hua, H., Liu, X., Zhang, Y., 2018. Human perturbation ofthe global phosphorus cycle: changes and consequences. Environ. Sci. Technol. 52(5), 2438–2450.

Zhang, G., Liu, H., Liu, R., Qu, J., 2009. Removal of phosphate fromwater by a Fe-Mn binaryoxide adsorbent. J. Colloid Interface Sci. 335 (2), 168–174.

Zhang, M., Gao, B., Yao, Y., Xue, Y., Inyang, M., 2012. Synthesis of porous MgO-biocharnanocomposites for removal of phosphate and nitrate from aqueous solutions.Chem. Eng. J. 210, 26–32.