This article was downloaded by: [University Library Utrecht]On: 04 October 2013, At: 06:42Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Environmental TechnologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tent20

Removal and recovery of nutrients as struvite fromanaerobic digestion residues of poultry manureY.D. Yilmazel a & G.N. Demirer aa Department of Environmental Engineering, Middle East Technical University, 06531 Ankara,TurkeyPublished online: 13 Jun 2011.

To cite this article: Y.D. Yilmazel & G.N. Demirer (2011) Removal and recovery of nutrients as struvite from anaerobicdigestion residues of poultry manure, Environmental Technology, 32:7, 783-794, DOI: 10.1080/09593330.2010.512925

To link to this article: http://dx.doi.org/10.1080/09593330.2010.512925

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Environmental Technology

Vol. 32, No. 7, May 2011, 783–794

ISSN 0959-3330 print/ISSN 1479-487X online© 2011 Taylor & FrancisDOI: 10.1080/09593330.2010.512925http://www.informaworld.com

Removal and recovery of nutrients as struvite from anaerobic digestion residues of poultry manure

Y.D. Yilmazel and G.N. Demirer*

Department of Environmental Engineering, Middle East Technical University, 06531 Ankara, Turkey

Taylor and Francis

(

Received 11 January 2010; Accepted 30 July 2010

)

10.1080/09593330.2010.512925

The removal and the recovery of nutrients, namely nitrogen (N) and phosphorus (P) from anaerobically digested andsolid–liquid separated manure effluents via struvite precipitation were investigated. Both the liquid and the solidphases of the poultry manure digester effluent were subjected to struvite precipitation experiments. The Mg:N:Pmolar ratio of 1:1:1 in the liquid phase resulted in an average NH

4

-N removal efficiency of 86.4%, which increasedto 97.4% by adjusting the Mg:N:P ratio to 1.5:1:1. The acidic phosphorus-dissolution process was applied to the solidphase of the effluent to obtain a phosphorus-enriched solution. Nutrient recovery experiments with NaOH as thebuffering reagent were conducted with and without addition of external chemicals (Mg and P sources) to evaluate theinfluence of the Mg:N:P molar ratio, the Mg:P molar ratio and pH. All the experiments depicted complete PO

4

-P(99.6–100.0%) and partial NH

4

-N (3.3–65.6%) recoveries from the phosphorus-enriched solution.

Keywords:

anaerobic digestion; poultry manure; nutrient recovery; phosphorus dissolution; struvite

Introduction

The poultry and the livestock industries are growingrapidly and this has led to large quantities of animalwaste production. Atuanya and Aigbirior [1] reportedthat the poultry production is the fastest growingcottage industry, and estimated the annual solid wastegeneration from poultry farms at the level of millionsof tons. Gungor-Demirci and Demirer [2] reported thatthe production of cattle and poultry manure in Turkeywas approximately 20 million tons of dry matter in2000. Direct land application of manure is the mostpreferred method of utilization, but it is not alwaysfeasible. Because nutrients in manure are not necessar-ily present in the same proportion needed by the crops,if applied at a rate higher than plant uptake, there is agreat risk of nutrient leaching and run-off resulting ineutrophication of surface waters [3]. Moreover, thehigh prevalence of pathogenic microorganisms in freshpoultry poses a potential threat to human healththrough contamination of water bodies from untreatedpoultry wastewater [4]. If the land to be used for directapplication is distant or the location is sensitive toodour, some type of manure treatment may be required.Some of the currently used options for the managementof poultry manure such as landfilling and incinerationlead to the loss of nutrients as well as environmentalproblems [5].

Anaerobic digestion (AD) is an established technol-ogy to convert animal waste into profitable by-productsas well as to reduce the relevant air, water and soilpollution problems [1,2,6,7]. However, because ADremoves mainly carbon, additional processes to removenitrogen (N) and phosphorus (P) should also be used tomeet the stringent effluent criteria. Moreover, there is ashift from the removal to the recovery of nutrients as aresult of increasing concerns regarding limited naturalresources and the importance given to the sustainabletreatment technologies.

Crystallization of N and P in the form of magnesiumammonium phosphate hexahydrate (MgNH

4

PO

4

·6H

2

O,struvite or MAP) is one of the possible techniquesused to remove and recover nutrients from wastewater[8–11]. Struvite is a valuable fertilizer since it releasesnutrients slowly and has non-burning features owing toits low solubility in water. Struvite formation isobserved at the stoichiometric ratio of 1:1:1 of the ionscomposing struvite according to the following reaction[10,12]:

The success of MAP precipitation depends on twomajor factors: Mg:N:P molar ratio and the pH of the

*Corresponding author. Email: goksel@metu.edu.tr

Mg NH PO H O

MgNH H O

24 4

32

4 2

6

6

+ + -+ + + fi

ŸPO (1)4

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3

784

Y.D. Yilmazel and G.N. Demirer

solution [13]. In a given solution, struvite can form andprecipitate if the product of Mg

2+

, NH

4+

and PO

43

−

ionactivities exceed the ion-activity product at equilibriumor the thermodynamic solubility product [12]. AlthoughH

+

is not directly involved in the reaction, struviteprecipitation is highly dependent on pH. This is simplydue to changes in the activities of both NH

4+

and PO

43

−

.Theoretically, solubility of struvite decreases as pHincreases up to a pH level of 9.0. At a pH level above9.0, struvite becomes more soluble as a result ofdecreasing and increasing ion activities of ammonia andphosphate, respectively [12]. However, the pH of mini-mum solubility of struvite may differ owing to changingionic strength and composition of wastewaters. Theminimum solubility pH for struvite precipitationreported by a number of researchers displays a range ofvalues, from 8.0 to 10.7 [10,12]. Struvite precipitation isalso influenced by chemical composition of the waste-water (organic matter, presence of chelating agents,ionic strength), the degree of supersaturation, tempera-ture and the presence of foreign ions such as calcium[8–10,14].

The struvite precipitation technique has beenapplied to various wastes, such as swine waste [15–17],dairy manure [11], calf manure [9], landfill leachate[18], semiconductor wastewater [19], slaughterhousewastewaters [20] and anaerobic digester sidestreams[21]. However, to the best of the authors’ knowledge,there is only one published work, by Yetilmezsoy andSapci-Zengin [22], on struvite precipitation from anaer-obically digested poultry manure. In their study, threecombinations of chemicals were tested at different pHlevels, and the performance of the struvite precipitationprocess was evaluated by measuring the remainingchemical oxygen demand (COD), colour and ammoniaconcentrations [23].

Therefore, in order to fill the gap in the literature,the current study aimed to recover N and P from bothphases of the anaerobically digested and solid–liquidseparated effluents of a full-scale poultry manure diges-tion plant. Most of the studies in the literature focusedon the removal/recovery of the readily available nutri-ents in the wastewaters [8,13,16,17]. However, thedissolution of nutrient from the solid phase is the onlyway to recover the highest amount of nutrients, particu-larly phosphorus, from animal wastes. Recently anumber researchers focused on the extraction of thenutrients present in the solid phase of the wastewatersludge [23,24], poultry litter [25] and piggery wastewa-ter [26]. This study illustrates the feasibility of therecovery of nutrients from the solid phase of anaerobi-cally digested poultry manure by the adoption of anovel phosphorus dissolution process, thereby obtain-ing a phosphorus-enriched liquid phase.

Materials and methods

Sample preparation

The wastewater sample was collected from the effluentof a full-scale biogas plant operated at mesophilicconditions. It digests poultry manure generated from asmall poultry farm housing 30,000 laying hens, locatedin Forchtenberg, Germany. The sample was kept refrig-erated at 4

°

C until used. Owing to the high solidscontent, the effluent sample was subjected to solid–liquid separation, and both phases of the effluent werecharacterized (Table 1).

Experimental set-up and procedures

The experimental study was performed at the laborato-ries of the Department of Wastewater Technology(AWT) at the University of Stuttgart, Germany.

Struvite precipitation experiments

The solid–liquid separation of the anaerobic digestereffluent was achieved by centrifugation (RC6, SorvallInstruments DuPont, Osterode, Germany) for 15minutes at 17,710

×

g

and sieving through a screen of0.56 mm (0.022 in) mesh size. The liquid phase wasused directly for struvite precipitation experiments,where the solid phase was subjected to the phosphorusdissolution process before the struvite precipitationexperiments.

Four consecutive steps were followed in the struviteprecipitation experiments: (1) addition of chemicals, (2)mixing, (3) settling, (4) filtration. The struvite precipi-tation experiments were conducted in continuouslystirred batch reactors at room temperature (21–22

°

C).Each struvite reactor contained 150 mL of the sample,and was mixed using a magnetic stirrer (MR Hei-Mix L,Heidolph, Schwabach, Germany). During the experi-ments mixing intensity was kept constant at 250 rpm. Inall experimental runs, there was no addition of ammonianitrogen (NH

4

-N). The total (initial + added) molarconcentration ratios of the ions were varied during theexperiments, and the concentrations of Mg and ortho-phosphate (PO

4

-P) were raised via addition of theseions externally. After adding the Mg-containing chemi-cal, the PO

4

-P-containing chemical was added to thereactor where necessary. The required amounts of Mgand PO

4

-P were calculated considering initial concen-trations of these ions in the wastewater, and were raisedaccordingly to obtain the desired total molar concentra-tion ratio of Mg:N:P in the reactor. In the experimentsMgCl

2

·6H

2

O was used in its solid form as the Mgsource and 75% H

3

PO

4

(v/v) was used as the P source.The pH adjustments were made using 20% NaOH (v/v)

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3

Environmental Technology

785

solution. Since the volume of NaOH added was verysmall, the dilution effect was negligible. All chemicalsused in the experiments were analytical grade. After thepH of the solution became constant at the desired levelwith a variation of

±

0.01 pH units, 30 minutes ofmixing was applied, and at the end of the mixing periodthe reactor content was allowed to settle down for 60minutes. After the struvite formation reaction wascompleted, the reactor content was filtered through afolded filter of pore size range 4–12

µ

m (MN615,Macherey-Nagel, Duren, Germany) and the filtrate wasanalyzed for its PO

4

-P, NH

4

-N and metal content. Theprecipitates were dried in a constant temperature roomat 30

±

2

°

C overnight. After drying, the precipitate wasseparated manually from the filter paper and kept atroom temperature (21–22

°

C) until analysed by X-raydiffraction (XRD).

Phosphorus dissolution process

The solid phase obtained after the solid–liquid separationof the biogas plant effluent was subjected to thephosphorus dissolution process prior to the struvite

precipitation experiments. Phosphorus dissolution wasachieved in four steps: (1) dilution, (2) acidic dissolution,(3) mixing and (4) solid–liquid separation (Figure 1). Thesolid phase was diluted by distilled water until the totalsolids (TS) concentration became 10%, which allowedcontinuous mixing in the reactor. Then, the pH of thediluted waste was adjusted to 2.0 by the addition of 20%HCl (v/v). The acidic mixture was mixed at 250 rpm byusing the jar test apparatus (MSR12/180, GeppertRuhrtechnik, Dreieich, Germany) for two hours. Afterthe dissolution, separation of the phosphorus-enrichedliquid phase from the remaining solid phase was achievedby centrifugation (RC6, Sorvall Instruments DuPont,Osterode, Germany) at 17,710

×

g

for 15 minutes andsieving through a screen of 0.56 mm (0.022 in) mesh size.

Figure 1. Shematic representation of the phosphorus dissolution process.

Liquid-phase experiments

Ammonium-nitrogen was in excess in the AD effluentand additions of Mg and PO

4

-P from external sourceswere necessary for struvite precipitation. This isbecause the molar ratio of struvite-forming ions(Mg:N:P ratio) should be at least equal for intentional

Table 1. Characterization of the solid–liquid separated effluent.

Concentration

Parameter Liquid phase

a

Unit Solid phase

a

Unit

TS 16.3

±

0.0 g kg

−

1

273

±

4 g kg

−

1

VS 57.4

±

1.9 % of TS 31.0

±

0.6 % of TSCOD 14,516

±

639 mg L

−

1

nd

b

–sCOD 3713

±

22 mg L

−

1

nd –TKN 5838

±

12 mg L

−

1

13

±

0.1 mg g

−

1

dry matterNH

4

-N 4612

±

117 mg L

−

1

nd –PO

4

-P 163

±

0 mg L

−

1

nd –TP 287

±

1 mg L

−

1

18.5

±

0.2 mg g

−

1

dry matterAl 1.39

±

0.03 mg L

−

1

703

±

35 mg kg

−

1

dry matterCa 78.6

±

6.90 mg L

−

1

21,6450

±

257 mg kg

−

1

dry matterCd < 0.025 mg L

−

1

< 2.8 mg kg

−

1

dry matterCo 0.07

±

0.00 mg L

−

1

< 1.7 mg kg

−

1

dry matterCr 0.08

±

0.01 mg L

−

1

20.9

±

0.0 mg kg

−

1

dry matterCu 0.46

±

0.00 mg L

−

1

41.6

±

0.6 mg kg

−

1

dry matterFe 7.24

±

0.08 mg L

−

1

1334

±

5 mg kg

−

1

dry matterHg <0.005 mg L

−

1

< 0.089 mg kg

−

1

dry matterK 3111

±

49 mg L

−

1

11,101

±

339 mg kg

−

1

dry matterMg 5.31

±

0.10 mg L

−

1

9037

±

357 mg kg

−

1

dry matterNi 0.21

±

0.00 mg L

−

1

12

±

0.8 mg kg

−

1

dry matterPb < 0.05 mg L

−

1

< 5.6 mg kg

−

1

dry matterZn 3.11

±

0.03 mg L−1 382 ± 0 mg kg−1 dry matterConductivity 32.2 mS cm−1 – –pH 8.58 – – –

Note: TS, total solids; VS, volatile solids; COD, chemical oxygen demand; SCOD, soluble COD; TKN, total Kjeldahl nitrogen; NH4–N, ammonia nitrogen; PO4–P, orthophosphate; TP, total phosphorous; aMean ± sd (n = 2); nd, not determined.

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3

786 Y.D. Yilmazel and G.N. Demirer

struvite precipitation. However, at any given pH level,any increase in the Mg:N:P ratio would increase thedegree of saturation with respect to the struvite forma-tion, which in turn would enhance the removal andrecovery of PO4-P and NH4-N [27,28].

The initial composition of the liquid phasecontained a molar concentration ratio of 1:1510:24 ofMg:N:P indicating that the concentrations of Mg andPO4-P ions were not adequate for higher removals ofNH4-N, and it was necessary to add them to attainhigher NH4-N removals. In order to observe the effectof the molar concentration ratio of the struvite-formingions, three different Mg:N:P ratios (1:1:1,1.3:1:1 and1.5:1:1) were used (Table 2). The molar concentrationratios were prepared by adding the calculated amount ofchemicals into the reactors, considering the initialconcentrations of the ions in the sample (Table 1).

During the Mg:N:P experiments the pH level was keptconstant at 8.5. To observe the effect of pH on theremovals of NH4-N and PO4-P, three different pH levels(8.0, 8.5 and 9.0) were investigated by keepingthe Mg:N:P ratio constant at 1:1:1 (Table 2). Theseexperiments were conducted keeping all other parame-ters the same. The experimental set-up is depicted inTable 2.

Solid-phase experiments

The characterization of the phosphorus-enriched liquidphase is depicted in Table 3. In the struvite precipita-tion experiments conducted with the phosphorus-enriched liquid phase, the effects of the Mg:N:P molarratio and the Mg:P molar ratio were investigated sepa-rately. In the experiments, three different Mg:N:P ratios

Figure 1. Shematic representation of the phosphorus dissolution process.Note: TS, total solids.

Table 2. The experimental set-up.

Experiment no Tested parameter Basis of chemical addition Molar ratio Chemicals added pH

Liquid-phase experimentsL1 pH NH4-N 1:1:1 H3PO4, MgCl2·6H2O 8.0L2 Mg:N:P, pH NH4-N 1:1:1 H3PO4, MgCl2·6H2O 8.5L3 pH NH4-N 1:1:1 H3PO4, MgCl2·6H2O 9.0L4 Mg:N:P NH4-N 1.3:1:1 H3PO4, MgCl2·6H2O 8.5L5 Mg:N:P NH4-N 1.5:1:1 H3PO4, MgCl2.6H2O 8.5

Solid-phase experimentsS1 Mg:N:P NH4-N 1:1:1 H3PO4, MgCl2·6H2O 8.5S2 Mg:N:P NH4-N 1.3:1:1 H3PO4, MgCl2·6H2O 8.5S3 Mg:N:P NH4-N 1.5:1:1 H3PO4, MgCl2·6H2O 8.5S4 Mg:P PO4-P 1:1 MgCl2·6H2O 8.5S5 Mg:P PO4-P 1.3:1 MgCl2·6H2O 8.5S6 Mg:P PO4-P 1.5:1 MgCl2·6H2O 8.5S7 _a _a 1:2b na 8.5S8 _a _a 1:2b na 9.5

Note: anot applicable; binitial Mg:P ratio; na, no addition

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3

Environmental Technology 787

(1:1:1,1.3:1:1 and 1.5:1:1) were used. The initial molarconcentration of the NH4-N (Table 4) was the highestamong the three struvite-forming ions and the initialMg:N:P molar ratio was 1:4:2. Therefore, the concen-tration of NH4-N was taken as the basis in order toadjust the desired Mg:N:P molar ratio (Table 2). TheMg:N:P molar ratio adjustments were performed asdescribed in the previous section.

To investigate the effect of the Mg:P molar ratio,three different Mg:P molar ratios, namely 1:1,1.3:1 and1.5:1, were used (Table 2). For the adjustment of theMg:P molar ratio, the initial PO4-P concentration in thesolution was taken as the basis (Table 3). The totalmolar concentration of Mg was raised according to theinitial molar concentration of PO4-P (57.4 ± 0.1 mM).

After the addition of the calculated amounts of Mg, thefinal Mg:N:P molar ratio became 1:2:1, 1.3:2:1, and1.5:2:1 for Experiment S4, S5 and S6, respectively.

To observe the effects of external Mg and PO4-Psources, two experiments (Experiment S7 and S8) wereconducted without addition of any chemical except thebuffering reagent (20% v/v NaOH). Sodium hydroxidewas added to the reactor in order to adjust the pH of thesolution to the desired value (8.5 and 9.5, in ExperimentS7 and S8, respectively).

Analytical methods

All the analyses (chemical oxygen demand: COD; solu-ble chemical oxygen demand: sCOD; volatilesuspended solids: VSS; total suspended solids: TSS;TS; NH4-N; total Kjeldahl nitrogen: TKN; total phos-phorus: TP; PO4-P and metals) were performed accord-ing to the Deutsches Institut fur Normung e.V. (DIN)standards [29]. All metal (Mg, Al, Ca, Cd, Co, Cr, Cu,Fe, Hg, K, Ni, Pb, Zn) analyses, except Mg of the liquidsamples, were carried out using an optical emissionspectrophotometer with inductively coupled plasma(ICP-OES) in the Hydrochemistry Department of theUniversity of Stuttgart. Magnesium measurements ofthe liquid samples were carried out using a Cadas 200Spectrophotometer (Dr. Lange, Dusseldorf, Germany)and LCK 326 reagent sets (Hach Lange, Dusseldorf,Germany).

XRD analysis

The XRD technique was used to determine the compo-sitional structure of the products and confirm the pres-ence of struvite in the precipitates. The dried precipitatewas separated manually from the filter paper and iden-tified by XRD. The XRD analyses of the precipitatescollected from L2, L5, S1 and S3 were performed at theAdvanced Analysis Laboratory (Ileri Analizler Labo-ratuvari) of Istanbul University, Turkey, using theRigaku D-Max 2000 X-ray diffractometer using Cu Kαradiation. The diffractometer was running at 40 kV and20 mA. The data were collected over the two–thetarange of 5–70° using step size of 0.05°, and counts werecollected for 1.5 seconds at each step. The XRD analy-ses and subsequent identification of the dried precipi-tates collected from Experiments S5, S6 and S8 wereconducted in the Karlsruhe Research Centre, Germany.The XRD analyses of the samples were conducted usingthe Siemens Bruker D5000 X-ray diffractometer usingCu Kα radiation by the Head of the Division of Nanom-ineralogy of the Karlsruhe Research Centre, Germany.The data were collected over the two–theta range of13.5–34.5° using step size of 0.03°, and counts werecollected for 10 seconds at each step.

Table 3. Characterization of the phosphorus-enriched liquidphase.

Parameter Concentration (mg L−1)

NH4-N 1969 ± 74a

PO4-P 1778 ± 2a

Mg 806 ± 1a

Al 4.06b

Ca 5152b

Cd <0.025b

Co <0.015Cr 0.208Cu 0.109b

Fe 37.9b

K 1057b

Ni 0.230b

Pb <0.05Zn 8.03b

Note: amean ± std (n = 2); bAl, Ca, Cd, Cu, Fe, K, Ni, Znconcentrations depict the mean concentration of the duplicateanalysis.

Table 4. Results of the liquid-phase experiments.

Residual concentration (mgL−1)a% removal/recoveryb

Experiment no NH4-N PO4-P Mg NH4-N PO4-P

L1 630 ± 4 3293 ± 1 10.6 ± 1.9 86.3 _L2 646 ± 35 3679 ± 1 5.65 ± 0.92 86.0 _L3 606 ± 0 3235 ± 1 9.55 ± 0.35 86.9 _L4 158 ± 0 1203 ± 4 45.5 ± 1.7 96.6 _L5 121 ± 3 112 ± 4 344 ± 7 97.4 31.6

Note: amean ± sd (n = 2); bAll removal/recovery efficiencies werecalculated considering the initial concentrations of the ions at theinfluent of the struvite reactor; no removal/recovery, i.e. finalconcentration is higher than initial concentration.

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3

788 Y.D. Yilmazel and G.N. Demirer

Results and discussion

In the liquid-phase experiments, precipitation led to the‘removal and recovery’ of the already existing nutrientsfrom the wastewater in the form of struvite. Therefore,the reductions in the concentrations of the ions arereferred to as ‘removal’. However, in the solid phaseexperiments, a pretreatment step was added to thesystem in order to make the nutrients available forstruvite reaction. Therefore, the reduction in theconcentrations of the nutrients from the phosphorus-enriched liquid phase is referred to as ‘recovery’ but notremoval.

Struvite precipitation from the liquid-phase effluent

The effect of the molar ratio

Experiment L1, L4 and L5 were performed at pH 8.5with different Mg:N:P molar ratios (Table 2) and theresults of the experiments are depicted in Table 4. Inorder to represent the actual removal of the ions fromthe wastewater, the removals are reported consideringthe initial concentrations of these ions present in waste-water (Table 1), but not the total (initial + added)concentrations.

The result of Experiment L2 indicated that theadjustment of the molar ratio of Mg:N:P to 1:1:1 at a pHlevel of 8.5 was not sufficient to decrease the residualconcentration of PO4-P (3679 ± 1 mg L−1) down to itsinitial concentration (163 ± 0 mg L−1, Table 1). Becauseof the PO4-P addition into the wastewater in order toinitiate the struvite formation, the residual concentra-tion was higher than the initial concentration. This ledto further contamination of the wastewater by PO4-P(Table 4). However, the removal of NH4-N moval was86.0% (reduced from 4612 ± 117 mg L−1 to 646 ± 35mg L−1). Based on the possible removal mechanisms ofNH4-N, this is attributable to the struvite formation[30]. In systems with high concentrations of Mg andPO4-P, there are two major mechanisms of NH4-Nremoval, namely struvite precipitation and air stripping.Even though struvite precipitation leads to formation ofother minerals containing Mg and/or PO4-P (e.g.hydroxylapatite, newberyite, monenite), examination ofall the possible minerals indicated that the only mineralcontaining NH4

+ in its composition is struvite [31].The effect of air stripping on NH4-N removal was

not determined in this study. However, the averageNH4-N removal via air stripping at pH 8.5 at roomtemperature (21–22°C) is recorded as 2% in anotherstudy, which used the effluent of a laboratory-scalepoultry manure digester [32]. This suggests that at a pHlevel of 8.5 there is negligible loss of ammonia to atmo-sphere. This finding contradicts the study of Celen andTurker [31], who observed 11% loss of ammonia to air

at pH 8.5. However, the difference can be explained bythe higher operational temperature (37°C), adjustedduring the experiments, used in [31], which is one of theimportant parameters increasing the rate of ammoniastripping [32,33]. Therefore, high removal efficienciesof NH4-N in the experiments (L1–L5) can be consid-ered as an indication of formation of struvite in the reac-tors. The XRD pattern generated from the samplematched the database standard for struvite (i.e. positionand intensity of the peaks, Figure 2), identifying theprecipitate as struvite, and no other minerals weredetected. On the other hand, the use of the Mg:N:Pmolar ratio of 1.5:1:1 (Experiment L5) increased NH4-N removal to 97.4% and reduced residual concentrationof PO4-P to 112 ± 4 mg L−1 resulting in a removal effi-ciency of 31.6% (Table 4). The requirement of excessMg for the enhancement of PO4-P removal can beexplained by the presence of the complexing agents inthe wastewater, which have the potential of formingsoluble complexes with Mg; this reduces the activity ofMg and makes it unavailable to the struvite reaction[16,34]. The precipitate produced in Experiment L5underwent XRD and was confirmed to be struvite.Figure 2. XRD patterns of the precipitate collected from Experiment L2. (The d-spacing of the strong lines of struvite are 4.2531, 5.5995, 5.8997, 2.9167, 2.6909, 2.6587, 4.1364, 2.7998, retrieved from powder diffraction (PDF) card 71-2089.)

The effect of pH

Another important factor to be considered in the struviteprecipitation reactor is pH because it has an effect onthe activities of the ions and also on the solubility ofstruvite. The final pH levels of the reactors wereadjusted to 8.0, 8.5 and 9.0. The maximum pH was setto 9.0 to avoid the loss of NH4

+ into the atmosphere byammonia stripping.

In the three experiments conducted at different pHlevels (Experiment L1, L2 and L3), NH4-N removals(86.3%, 86.0% and 86.9%) were similar (Table 4). Asthe solution becomes more saturated with respect tothe ions forming struvite, the change in free ionconcentrations becomes less influential in the pHrange studied, thereby reducing the effect of pH on theremoval of NH4-N. A similar finding was recorded byUludag-Demirer and Othman [30], who also studiedthe effect of the molar ratio at pH levels of 8.0, 8.5and 9.0.

Struvite precipitation from the solid-phase effluent

The recovery of N and P via struvite precipitation canonly be achieved by using the dissolved fraction; there-fore, the solid phase was subjected to the phosphorusdissolution process (Figure 1). Muller et al. [35]reported that sludge cells can be dissolved by acidic oralkali treatment at low or ambient temperatures. In thestudy of Weidelener et al. [23], acidic and alkalinedissolution processes were compared for three different

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3

Environmental Technology 789

sewage sludge samples, and it was reported that theefficiency of the acidic dissolution was higher than thatof the alkaline dissolution. Moreover, Szogi et al. [25]successfully applied acidic treatment for the purpose ofP extraction from poultry litter. In a recent study, acidicdissolution was employed for the recovery of P fromash of incinerated sewage sludge and animal carcassesand enabled more than 85% P dissolution [36]. Basedon these, acidic dissolution was used in this study(Figure 1).

The solid phase obtained after solid–liquid separa-tion was rich in nutrients (Table 1). The applied processled to the transformation of available phosphates intoorthophosphate (Table 3). This can be attributable to thedisintegration of the cell walls and mineralization of themicrobial cells; similar results were reported in otherstudies [22,36]. This process also releases P from insol-uble inorganic phosphate complexes [25]. In addition tothe increase in PO4-P concentration, acidic dissolutionresulted in the release of the metals (Table 3) that werenormally integrated in organic complex molecules intothe liquid phase [37]. The phosphorus-enriched liquidphase was analysed for its metals/heavy metals(Table 3) because they can be incorporated into thecrystal lattice or sorbed into the surface of struvite andcan reduce the purity of the struvite in the precipitatingmineral [38].

The effect of the Mg:N:P molar ratio

Experiment S1, S2 and S3 were performed in order toinvestigate the effect of Mg:N:P molar ratio (Table 4).The results of Experiment S1 illustrated that the use ofthe Mg:N:P molar ratio of 1:1:1 was sufficient forcomplete PO4-P recovery, reducing its concentrationfrom 1778 ± 2 mg L−1 (Table 3) to 0.88 ± 0.03 mg L−1

(Table 5). However, there was only partial recovery(61.5%) of NH4-N in the experiment. The differencebetween the recovery efficiencies of NH4-N and PO4-Pmay be attributable to the formation of several insolublespecies together with struvite [30]. Therefore, the XRDanalysis of the precipitate collected from Experiment S1and S3 were conducted. The XRD patterns of theprecipitates collected from both experiments werematched with struvite (Figure 3 and Figure 4, for Exper-iment S1 and Experiment S3, respectively). However,there were indications of the presence of poorly crystal-lized materials (a broad hump centred at approximately30° in the XRD pattern); this explains the completerecovery of PO4-P from the solution via formation ofother P-containing minerals.Figure 3. XRD patterns of the precipitate collected from Experiment S1. (The d-spacing of the strong lines of struvite are 4.2531, 5.5995, 5.8997, 2.9167, 2.6909, 2.6587, 4.1364, 2.7998, retrieved from PDF card 71-2089. The d-spacing of the strong lines of hydroxylapatite are 2.8147,2.7205, 2.7781, 3.4395, 1.8403, 1.9437, 2.6299, 2.2636, retrieved from PDF card 74-0566.)Figure 4. XRD analysis of the precipitate collected from th Experiment S3. (The d-spacing of the strong lines of struvite are 4.2531, 5.5995, 5.8997, 2.9167, 2.6909, 2.6587, 4.1364, 2.7998, retrieved from PDF card 71-2089. The d-spacing of the strong lines of hydroxylapatite are 2.8147,2.7205, 2.7781, 3.4395, 1.8403, 1.9437, 2.6299, 2.2636, retrieved from PDF card 74-0566.)In the literature, it is stated that Ca is one of theimportant ions interfering with the precipitation of stru-vite and the removal of PO4-P [14,21,39,40]. The effectof the Ca ion has been studied by several investigators

Figure 2. X-ray diffraction patterns of the precipitate collected from Experiment L2. (The d-spacing of the strong lines of stru-vite are 4.2531, 5.5995, 5.8997, 2.9167, 2.6909, 2.6587, 4.1364, 2.7998, retrieved from powder diffraction card 71-2089.)

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3

790 Y.D. Yilmazel and G.N. Demirer

Figure 3. X-ray diffraction patterns of the precipitate collected from Experiment S1. (The d-spacing of the strong lines ofstruvite are 4.2531, 5.5995, 5.8997, 2.9167, 2.6909, 2.6587, 4.1364, 2.7998, retrieved from power diffraction card 71-2089. Thed-spacing of the strong lines of hydroxylapatite are 2.8147, 2.7205, 2.7781, 3.4395, 1.8403, 1.9437, 2.6299, 2.2636, retrievedfrom power diffraction card 74-0566.)

Figure 4. X-ray diffraction analysis of the precipitate collected from th Experiment S3. (The d-spacing of the strong lines ofstruvite are 4.2531, 5.5995, 5.8997, 2.9167, 2.6909, 2.6587, 4.1364, 2.7998, retrieved from power diffraction card 71-2089. Thed-spacing of the strong lines of hydroxylapatite are 2.8147, 2.7205, 2.7781, 3.4395, 1.8403, 1.9437, 2.6299, 2.2636, retrievedfrom power diffraction card 74-0566.)

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3

Environmental Technology 791

to enhance P removal, and it is postulated that the rela-tive concentrations of Mg and Ca are the major factorsdetermining P removal performance and degree of stru-vite formation [39,40]. The relative concentration of Cato Mg was 0.9 in Experiment S1, whereas the same ratiowas 0.6 in Experiment S3. The presence of a highconcentration of Ca ions limited the recovery of NH4-Nand struvite crystallization [40,41].

The trend followed for the recovery of PO4-P andNH4-N in Experiment S1, S2 and S3 was the same withthe liquid-phase Mg:N:P experiments (L2, L4 and L5).The highest percentage removals/recoveries wereattained by the use of a Mg:N:P ratio of 1.5:1:1. Thedifferences between the efficiencies can be attributableto the different chemical compositions (nutrient concen-trations, ionic strength, Ca and other metal concentra-tions) of the samples investigated.

The effect of the Mg:P molar ratio

The results of the Mg:P experiments are depicted inTable 5. The concentrations of Cd, Cu, Co, Pb, Cr andNi in the remaining solution were not determined asthey were already in trace concentrations in the originalsolution (Table 3).

The results of Experiment S6, S5 and S4 revealedthat residual PO4-P concentrations ranging between1.00 ± 0.04 mg L−1 and 1.20 ± 0.03 mg L−1 were achiev-

able by the addition of Mg only. However, since theresidual concentrations of NH4-N (1885 ± 6 mg L−1 to1844 ± 2 mg L−1) were quite high, there should be post-treatment of this ion downstream to meet the dischargeregulations. The effluent discharge criteria specified bythe World Bank Group for poultry processing is 10 mgL−1 for total N and 2 mg L−1 for total P [42], and thesimilar criteria required by the Water Pollution ControlRegulation of Turkey are 15 mg L−1 for NH4-N and 2mg L−1 for PO4-P [43].

Comparison of Experiment S4 and Experiment S5indicated that the addition of Mg at a concentrationhigher than the stoichiometric ratio of PO4-P loweredthe recovery efficiency of NH4-N (Table 5). The highCa recovery efficiencies observed in Experiment S5, S6and S7, when considered together with the high initialconcentration of Ca of 5152 mg L−1 (Table 5) indicatedthe formation of calcium-containing minerals in thereactor. To determine the precipitating mineral, XRDanalyses were conducted for the precipitates collectedfrom Experiments S5 and S6. These XRD patternsmatched the patterns of struvite, and also there wereindications of the presence of amorphous phase miner-als (data not shown). The presence of amorphouscalcium-containing minerals together with struviteconfirmed the presence of alternative mechanisms forremoval of PO4-P from the system, leading to completerecovery of PO4-P.

Table 5. Results of the solid-phase experiments.

Residual concentration (mg L−1)a

Experiment no. NH4-N PO4-P Mg Al Ca Fe K Zn

S1 760 ± 2 0.88 ± 0.03 621 ± 27 nd nd nd nd ndS2 720 ± 1 0.80 ± 0.05 1325 ± 49 nd nd nd nd ndS3 678 ± 4 0.57 ± 0.04 1738 ± 4 nd nd nd nd ndS4 1844 ± 2 1.20 ± 0.03 1023 ± 15 nd nd nd nd ndS5 1885 ± 6 1.03 ± 0.01 1244 ± 7 0.20 1944 ± 21 <0.09 1056 ± 16 ndS6 1878 ± 4 1.00 ± 0.04 1229 ± 4 nd 1995 ± 7 nd 1039 ± 6 ndS7 1906 ± 11 7.90 ± 0.00 598 ± 25 0.14 1310 ± 10 0.26 ± 0.03 757 ± 2 0.19 ± 0.01S8 1640 ± 36 0.11 ± 0.00 710 ± 7 0.22 nd 0.07 1036 0.87

Experiment no. Recovery (%)b

NH4-N PO4-P Mg Al Ca Fe K ZnS1 61.5 100.0 22.9 nd nd nd nd ndS2 63.5 100.0 – nd nd nd nd ndS3 65.6 100.0 – nd nd nd nd ndS4 6.5 99.9 – nd nd nd nd ndS5 4.4 99.9 – 95.0 62.3 >99.8 0.1 ndS6 4.8 99.9 – nd 61.3 nd 1.7 ndS7 3.3 99.6 25.8 96.6 74.6 99.3 28.4 97.6S8 16.9 100.0 11.9 94.6 nd 99.8 2.0 89.2

Notes: amean ± sd (n = 2); ball recovery efficiencies were calculated considering the initial concentrations of the ions at the influent of struvitereactor (Table 3); nd, not determined; – , no recovery, i.e. final concentration is higher than initial concentration.

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3

792 Y.D. Yilmazel and G.N. Demirer

Experiments S7 and S8 were conducted withoutaddition of any chemical except the buffering reagentused to increase the pH to 8.5 and 9.5, respectively. TheXRD analysis of the precipitate collected from Experi-ment S8 depicted that there was a small amount of crys-talline material formed in the reactor, and most of theprecipitate was in the amorphous form (data notshown). The peaks that are formed during the analysisof the precipitate collected from Experiment S8 indi-cated the formation of hydroxlyapatite (HAP,Ca5(PO4)3(OH)) and struvite. According to the literaturethere are five calcium phosphate crystalline species thatcan precipitate from solutions containing Ca and P.These are HAP, tricalcium phosphate (whitlockite)(TCP, Ca3(PO4)2), octacalcium phosphate (OCP,Ca8(HPO4)2(PO4)4·5H2O), monenite (DCP, CaHPO4)and dicalcium phosphate dihydrate (brushite) (DCPD,CaHPO4·2H2O) [13,23,30,33 ]. Among them, the mostthermodynamically stable phase is HAP, which couldbe expected to precipitate [12,44]. However, it has beenestablished that a number of species act as precursors tothe precipitation of HAP, such as amorphous calciumphosphate (ACP, with approximate formulationCa3(PO4)2·xH2O), but with no structured crystallineorder [44,45], octacalcium and brushite. With timethese species may transform to HAP [33]. However, asreported in the study of Musvoto et al. [44], the forma-tion of the precursor species is a relatively fast process,whereas the growth of HAP is very slow, such that theconversion takes a long time. Moreover, the presence ofMg in solution strongly affects the conversion process[46]. Based on these discussions it may be speculatedthat there is formation of ACP in the reactors, which actas a precursor of HAP formation. However, the trans-formation of ACP to HAP did not occur in the time thatthe analysis was conducted.

Although the residual NH4-N concentration inExperiment S8 is lower than the level in S7, the residualMg concentration in Experiment S8 is higher incomparison to S7. These results depicted the loss ofammonia to atmosphere in Experiment S8 due to anelevated ammonia stripping rate at pH 9.5. Comparisonof Experiment S7 with the experiments conducted bythe addition of Mg at pH 8.5 indicated the effect ofaddition of Mg. The Ca recovery efficiency recorded inExperiment S7 was 74.6%, whereas the Ca recoveryefficiencies recorded in experiments conducted with theaddition of Mg at a concentration higher than the molarconcentration of initial PO4-P were 62.3% and 61.3% inExperiment S5 and S6, respectively. This indicated thatthe increase in Mg concentration in the solutiondecreased the precipitation of Ca-containing minerals.This finding is supported by other studies reporting thatthe magnesium ion kinetically hinders the nucleationand subsequent growth of HAP [17] and OCP [17,47].

Table 5 depicts the recovery efficiencies of the otherions (Al, Fe, K and Zn). The high recovery efficienciesof Al, Fe and Zn indicated their precipitation togetherwith struvite. A similar observation was recorded in thestudy of Rontentalp et al. [38], which stated that themetals can be incorporated into the crystal lattice orsorbed to the surface of struvite, decreasing the purity ofthe product. However, because the molar concentrationratios of Al:P, Fe:P and Zn:P were 0.003:1, 0.03:1 and0.002:1, they did not lead to formation of significantquantities of other minerals, such as berlinite (AlPO4),iron phosphate (FePO4) and zinc ammonium phosphatehexahydrate (ZnNH4PO4·6H2O), that were detectableby XRD analysis. The low participation of potassium inadsorption or precipitation reactions in wastewater andsludge treatment has been previously stated [48]. One ofthe possible species to be formed under these conditionsis potassium struvite (KMP, KMgPO4·6H2O). In theliterature, different authors [9,49] have pointed out thatpotassium struvite could precipitate instead of ammo-nium struvite only in the case of low ammoniumconcentrations. However, depending on the initialpotassium concentration, the co-precipitation of K-struvite and struvite is possible, although in smalleramounts, even with ammonia concentrations above2000 mg L−1 [50]. Based on this, it can be speculatedthat the relatively low residual concentration of potas-sium in Experiment S7, in comparison to others, mightbe due to the co-precipitation of K-struvite and struvitein the reactor. However, there is no data supporting this,because the amount of precipitate was not enough forXRD analysis.

Conclusions

The effluent from a mesophilic poultry manure digesterwas used in this study mainly because it is one of therichest wastes in terms of nutrients (N and P) and isproduced in high quantities all around the world. Theexperimental set-up was designed to attain the highestnutrient recovery potential from the waste by utilizingboth the liquid and the solid phase of the anaerobicallydigested poultry manure. The struvite precipitationexperiments were conducted under varying conditionsof Mg:N:P molar ratio, pH and Mg:P molar ratio. Theresults obtained depicted the following.

● The effect of pH in the liquid phase was negligi-ble in terms of NH4-N recovery efficiency. Thesimulation of struvite stoichiometry with excessMg (Mg:N:P ratio of 1.5:1:1) was enough foralmost complete (97.4%) recovery of NH4-N.

● The acidic dissolution process can be usedsuccessfully as a preliminary step of struviteprecipitation to obtain a nutrient-rich solution

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3

Environmental Technology

793

from the solid phase of digested animal manure.However, the process led to the dissolution ofmetals along with the nutrients, which led to thecrystallization of other minerals together withstruvite.

●

The presence of high concentrations of Ca in thephosphorus-enriched liquid phase led to completerecovery of PO

4

-P from the solution by theprecipitation of calcium phosphates together withstruvite.

●

The addition of an external P source was neces-sary for the recovery of NH

4

-N along with thePO

4

-P from the phosphorus-enriched solution.

Acknowledgements

Y.D. Yilmazel was a visiting scientist at the Institute forSanitary Engineering, Water Quality and Solid WasteManagement (ISWA), University of Stuttgart, Germanyduring the time of the study. The authors would like to thankthe members of the Institute for Sanitary Engineering, WaterQuality and Solid Waste Management (ISWA), University ofStuttgart, Germany, for technical support. This study wasfunded by the Scientific and Technological Council of Turkeythrough Grant Number 107Y231 and the BMBF (Germany)through the Intensified Cooperation (IntenC): Promotion ofGerman-Turkish Higher Education Research Program.

References

[1] E.I. Atuany and M. Aigbirior,

Mesophilic biomethana-tion and treatment of poultry wastewater using pilotscale UASB reactor,

Environ. Monit. Assess. 77 (2002),pp. 139–147.

[2] G. Gungor-Demirci and G.N. Demirer,

Effect of initialCOD concentration, nutrient addition, temperature andmicrobial acclimation on anaerobic treatability ofbroiler and cattle manure,

Bioresour. Technol. 93(2004), pp. 109–117.

[3] C. Burton and C. Turner,

Manure management treat-ment strategies for sustainable agriculture,

SilsoeResearch Institute, Silsoe, UK, 2003.

[4] M. Ghosh, A. Ganguli, and S. Pathak,

Application of anovel biopolymer for removal of

Salmonella

from poultrywastewater,

Environ. Technol. 30 (2009), pp. 337–344.[5] R. James, M.L. Eastridge, L.C. Brown, K.H. Elder, S.S.

Foster, J.J. Hoorman, M.J. Joyce, H.M. Keener, K.Mancl, M.J. Monnin, J.N. Rausch, J.M. Smith, O.Tuovinen, M.E. Watson, M.H. Wicks, N. Widman, andL. Zhao,

Ohio Livestock Manure Management Guide,

Ohio State University Extension Bulletin 604-06, 2006.Available at http://ohioline.osu.edu/b604/pdf/b604.pdf.

[6] C. Aubart and S. Fauchille,

Anaerobic digestion ofpoultry wastes – part 1. Biogas production and pollu-tion decrease in terms of retention time and total solidscontent,

Process Biochem. 2 (1983), pp. 31–34.[7] P. Mähnert and B. Linke,

Kinetic study of biogasproduction from energy crops and animal waste slurry:Effect of organic loading rate and reactor size,

Environ.Technol. 30 (2009), pp. 93–99.

[8] S. Uludag-Demirer,

A study on nutrient removal frommunicipal wastewater by struvite formation using Tagu-

chi’s design of experiments,

Environ. Eng. Sci. 25(2008), pp. 1–10.

[9] R.D. Schuiling and A. Andrade,

Recovery of struvitefrom calf manure,

Environ. Technol. 20 (1999),pp. 765–769.

[10] J.D. Doyle and S.A. Parsons,

Struvite formation, controland recovery,

Water Res. 36 (2002), pp. 3925–3940.[11] S. Uludag-Demirer, G.N. Demirer, and S. Chen,

Ammo-nia removal from anaerobically digested dairy manureby struvite precipitation,

Process Biochem. 40 (2005),pp. 3667–3674.

[12] V.L. Snoeyink and D. Jenkins,

Water Chemistry,

JohnWiley, New York, 1980.

[13] E.V. Munch and K. Barr,

Controlled struvite crystalliza-tion for removing phosphorus from anaerobic digestersidestreams,

Water Res. 9 (2001), pp. 151–159.[14] K.S. Le Corre, E. Valsami-Jones, P. Hobbs, and S.A.

Parsons,

Impact of calcium on struvite crystal size, shapeand purity,

J. Cryst. Growth 283 (2005), pp. 514–522.[15] L.J. Beal, R.T. Burns, and K.J. Stalder,

Effect of anaer-obic digestion on struvite production for nutrientremoval from swine waste prior to land application,

ASAE Annual International Meeting, Paper No.994042, St. Joseph, MI, 1999.

[16] N.O. Nelson, R.E. Mikkelsen, and D.L. Hesterberg,

Struvite precipitation in anaerobic swine lagoon liquid:Effect of pH and Mg:P ratio and determination of rateconstant,

Bioresour. Technol. 89 (2003), pp. 229–236.[17] I. Celen, J.R. Buchanan, R.T. Burns, R.B. Robinson,

and D.R. Raman,

Using a chemical equilibrium modelto predict amendments required to precipitate phospho-rus as struvite in liquid swine manure,

Water Res. 41(2007), pp. 1689–1696.

[18] I. Kabdasli, A. Safak, and O. Tunay,

Bench-scale eval-uation of treatment schemes incorporating struviteprecipitation for young landfill leachate,

WasteManage. 28 (2008), pp. 2386–2392.

[19] B. Warmadewanthi and J.C. Liu,

Selective separationof phosphate and fluoride from semiconductor waste-water,

Water Sci. Technol. 59 (2009), pp. 2047–2053.[20] I. Kabdasli, O. Tunay, and P. Ozcan,

Application ofstruvite precipitation coupled with biological treatmentto slaughterhouse wastewaters,

Environ. Technol. 30(2009), pp. 1095–1101.

[21] P. Battistoni, P. Pavan, M. Prisciandaro, and F. Cecchi,

Struvite crystallization: A feasible and reliable way tofix phosphorus in anaerobic supernatants,

Water Res.34 (2000), pp. 3033–3041.

[22] K. Yetilmezsoy and Z. Sapci-Zengin,

Recovery ofammonium nitrogen from the effluent of UASB treatingpoultry manure wastewater by MAP precipitation as aslow release fertilizer,

J. Hazard. Mater. 167 (2009),pp. 260–269.

[23] A. Weidelener, K. Brechtel, W. Maier, J. Krampe, andU. Rott,

Recovery of phosphorus from sewage sludge asMAP,

IWA/WISA Conference on the Management ofResidues Emanating from Water and Wastewater Treat-ment, Johannesburg, 2005.

[24] K. Guney, A. Weidelener, and J. Krampe,

Phosphorusrecovery from digested sewage sludge as MAP by thehelp of metal ion separation,

Water Res. 42 (2008),pp. 4692–4698.

[25] A.A. Szogi, M.B. Vanotti, and P.G. Hunt,

Phosphorusrecovery from poultry litter,

Trans. ASABE 51 (2008),pp. 1727–1734.

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3

794 Y.D. Yilmazel and G.N. Demirer

[26] M.L. Daumer, F. Beline, and S.A. Parsons, Chemicalrecycling of phosphorus from piggery wastewater, Inter-national Conference on Nutrient Recovery, Vancouver,Canada, 2009.

[27] A. Adnan, M. Dastur, D.S. Mavinic, and F.A. Koch,Preliminary investigation into factors affectingcontrolled struvite crystallization at the bench scale, J.Environ. Eng. Sci. 3 (2004), pp. 195–202.

[28] M.S. Rahaman, N. Ellis, and D.S. Mavinic, Effects ofvarious process parameters on struvite precipitationkinetics and subsequent determination of rateconstants, Water Sci. Technol. 57 (2008), pp. 647–654.

[29] Deutsches Institut fur Normung e.V. (DIN), DeutscheEinheitsverfahren zur Wasser, Abwasser und Schlam-muntersuchung – Allegemeine Angaloen (Gruppe A).DIN 38402-11

[30] S. Uludag-Demirer and M. Othman, Removal ofammonium and phosphate from the supernatant ofanaerobically digested waste activated sludge bychemical precipitation, Bioresour. Technol. 100(2009), pp. 3236–3244.

[31] I. Celen and M. Turker, Recovery of ammonia fromanaerobic digester effluents, Environ. Technol. 22(2001), pp. 1263–1272.

[32] Metcalf and Eddy, Inc., G. Tchobanoglous, F.L. Burton,and H.D. Stensel, Wastewater Engineering: Treatment,Disposal and Reuse, McGraw-Hill, New York, 2003.

[33] Y.D. Yilmazel, Removal and recovery of nutrients asstruvite from anaerobic digestion/co-digestion residuesof poultry manure, M.Sc. Thesis, Middle East Techni-cal University, 2009.

[34] R.T Burns, L.B. Moody, I. Celen, and J.R. Buchanan,Optimization of phosphorus precipitation from swinemanure slurries to enhance recovery, Water Sci.Technol. 48 (2003), pp. 139–146.

[35] J.A. Muller, A. Winter, and G. Stunkmann, Investiga-tion and assessment of sludge pre-treatment processes,Water Sci. Technol. 49 (2004), pp. 97–104.

[36] Y. Cohen, Phosphorus dissolution from ash of inciner-ated sewage sludge and animal carcasses using sulphu-ric acid, Environ. Technol. 30 (2009), pp. 1215–1226.

[37] E. Neyens, J. Baeyens, M. Weemas, and B. De Heyder,Hot acid hydrolysis as a potential treatment of thickenedsewage sludge, J. Hazard. Mater. 98 (2003), pp. 275–293.

[38] M. Rontentalp, M. Mauer, and W. Gujer, The behav-iour of pharmaceuticals and heavy metals during

struvite precipitation in urine, Water Res. 41 (2007),pp. 1859–1868.

[39] G.A Momberg and R.A. Oellermann, The removal ofphosphate by hydroxyapatite and struvite crystalliza-tion in South Africa, Water Sci. Technol. 26 (1992),pp. 987–996.

[40] J. Wang, J.G. Burken, X.J. Zhang and R. Suampalli,Engineered struvite precipitation: Impacts of compo-nent-ion molar ratios and pH, J. Environ. Eng. 131(2005), pp. 1433–1440.

[41] K.S. Le Corre, E. Valsami-Jones, P. Hobbs, and S.A.Parsons, Impact of reactor operation on success ofstruvite precipitation from synthetic liquor, Environ.Technol. 28 (2007), pp. 1245–1256.

[42] International Finance Corporation (IFC), Environmen-tal, Health, and Safety Guidelines for Poultry Process-ing, 2005. Available at http://www.ifc.org/ifcext/enviro.nsf/AttachmentsByTitle/gui_EHSGuidelines2007_PoultryProcessing/$FILE/Final+-+Poul-try+Processing.pdf.

[43] Turkish Ministry of Environment and Forestry (MoEF),Water Pollution Control Regulation, Official Gazetteno. 25687, 31 December 2004.

[44] E.V. Musvoto, M.C. Wentzel, and G.A. Ekama, Inte-grated chemical-physical processes modelling – II:Simulating aeration treatment of anaerobicdigester supernatants, Water Res. 34 (2000), pp.1857–1867.

[45] N.C. Blumenthal, F. Betts, and A.S. Posner, Stabilisa-tion of amorphous calcium phosphate by Mg and ATP,Calcif. Tissue Res. 23 (1977), pp. 245–250.

[46] E. Arvin, Observations supporting phosphate removalby biologically mediated chemical precipitation – areview, Water Sci. Technol. 15 (1983), pp. 43–63.

[47] F. Abbona, Crystallization of calcium and magnesiumphosphates from solutions of low concentration, J.Cryst. Growth 104 (1990), pp. 661–671.

[48] N. Marti, A. Bouzas, A. Seco, and J. Ferrer, Struviteprecipitation assessment in anaerobic digestionprocesses, Chem. Eng. J. 14 (2008), pp. 67–74.

[49] J. Wilsenach and M. Van Loosdrecht, Integration ofprocesses to treat wastewater and source-separatedurine, J. Environ. Eng. 132 (2006), pp. 331–341.

[50] L. Zeng and X. Li, Nutrient removal from anaerobicallydigested cattle manure by struvite precipitation, J.Environ. Eng. 5 (2006), pp. 285–294.

Dow

nloa

ded

by [

Uni

vers

ity L

ibra

ry U

trec

ht]

at 0

6:42

04

Oct

ober

201

3