- Email: [email protected]
brackish aquaculture condition
Bioresource Technology 218 (2016) 1151–1156
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Struvite crystallization under a marine/brackish aquaculture condition Xuedong Zhang ⇑, Jianmei Hu, Henri Spanjers, Jules B. van Lier Section Sanitary Engineering, Department of Water Management, Delft University of Technology, Stevinweg 1, 2628CN Delft, The Netherlands
h i g h l i g h t s ) and DH0r (25.7 kJ mol1) of struvite under the saline condition were obtained. 2+ and release of NH4+ led to low PO3 4 in the digester. Minimal solubility of struvite under marine/brackish conditions was at pH around 10. Average crystal size of struvite under the conditions decreased with pH. + High NH4 enhanced formation of ammonia precipitates, e.g. struvite and dittmarite. o
K sp (10
13.06
Natural presence of Mg
a r t i c l e
i n f o
Article history: Received 25 May 2016 Received in revised form 19 July 2016 Accepted 20 July 2016 Available online 21 July 2016 Keywords: Struvite crystallization Marine/brackish aquaculture condition Particle size distribution Thermodynamic solubility product Standard enthalpy
a b s t r a c t The results in this study show that struvite was formed in the digester at pH 7.7 due to the magnesium naturally present and the released ammonia and phosphate, resulting in low phosphate concentration in the digester. Apparently the digester already provided proper conditions for struvite formation. Under the brackish condition, the estimated thermodynamic solubility product and enthalpy change of struvite formation were 1013.06 and 25.7 kJ mol1, respectively. The average crystal size under marine/brackish condition decreased with pH, but increased with temperature. X-ray diffraction measurements indicate struvite (NH4MgPO46H2O) and dittmarite (NH4MgPO4H2O) were predominant phosphorus species produced in filtrates of the digester. However, struvite and newberyite (HMgPO43H2O) were the predominant species precipitated from synthetic brackish waters after dosing MgCl2. It is pronounced that (waste)water characteristics played also an important role on the nature of phosphate precipitates. Under high NH+4 condition, phosphorus precipitates containing ammonia were dominant, compared to other amorphous phosphates. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Nowadays phosphorus ending up in waste streams such as effluents from recirculation aquaculture systems (RAS) is a concern because the emission of phosphorus causes eutrophication of local water bodies. Meanwhile, steady supply of P fertilizer is essential for sustainable agricultural production to feed the soaring population (Cordell et al., 2009). In recent years there is an increasing concern about the exhaustion of exploitable phosphorus reserves (Vaccari, 2009; Xu et al., 2015), leading to a decline in phosphate rock production in some countries and finally resulting in the soaring price of phosphate rock. Due to the aforementioned issues, there is an increasing attention to investigation of struvite (NH4MgPO46H2O) precipitation and thereby recovery of phosphorus from various types of P rich waste streams, such as urine and ⇑ Corresponding author. E-mail addresses: [email protected], [email protected] (X. Zhang). http://dx.doi.org/10.1016/j.biortech.2016.07.088 0960-8524/Ó 2016 Elsevier Ltd. All rights reserved.
rejected water from digesters (He et al., 2016; Lahav et al., 2013; Liu et al., 2013; O’Neal and Boyer, 2013; Ronteltap et al., 2007, 2010). However, to date no report on struvite crystallization/precipitation under an aquaculture brackish condition is available yet, although there is a potential to recover P from P-rich sludge from aquaculture recirculation systems. Most studies on struvite crystallization/precipitation are carried out under low ionic strength conditions. Several studies on struvite precipitation under high ionic strength are available (Bhuiyan et al., 2007; Crutchik and Garrido, 2011; Crutchik et al., 2013; Liu et al., 2013; O’Neal and Boyer, 2013; Ronteltap et al., 2007, 2010). The ionic strength influences activity coefficients of ions, particularly the multivalent ions, such as Mg2+ and PO3 4 , which may further affect the thermodynamic parameters of struvite formation, such as the thermodynamic solubility product and the enthalpy of struvite formation. Theoretically, under high ionic strength conditions, struvite solubility increases due to the decreased activity coefficients of Mg2+, NH+4 and PO3 4 . However,
1152
X. Zhang et al. / Bioresource Technology 218 (2016) 1151–1156
limited literature on struvite behavior under high ionic strength conditions, i.e. marine/brackish conditions, is available, and thus further investigation on struvite formation under the high salinity condition is needed. In marine/brackish RAS, spontaneous struvite formation could occur, due to the naturally present Mg in the marine/brackish culturing water, and ammonia and phosphate released from degradation of fish faeces and/or fish feed (Zhang et al., 2013, 2016). Therefore, uncontrolled struvite scaling might become a problem for the piping system of marine/brackish RAS. Moreover, when anaerobic digestion for sludge stabilization is applied, the production of a P- and N- rich stream in and from the digestate will further enhance struvite precipitation. Furthermore, the digestate from a digester with controlled struvite precipitation in treating the salty sludge could further serve as fertilizer for salty plants. Therefore, a better understanding of struvite precipitation under marine/brackish aquaculture conditions is conducive to avoiding struvite scaling and to better unveiling struvite precipitation in the anaerobic digestion process treating saline (waste)waters. It is commonly regarded that pH and supersaturation have major impacts on struvite precipitation (Hanhoun et al., 2011). In addition, temperature as well affects struvite solubility and the enthalpy change of struvite formation (Bhuiyan et al., 2007; Hanhoun et al., 2011). Thus, the effects of pH, supersaturation, and temperature on struvite precipitation under marine/brackish aquaculture conditions are of great interest to be investigated. This study aims to understand struvite precipitation in digestates from an anaerobic digester treating the salty sludge from a marine/brackish RAS. Thus, in order to simplify the analyses in the batch tests, artificial mimicked supernatants of digestate were employed, and precipitation of the precipitates, such as calcium phosphate, calcium carbonate, and magnesium carbonate, was not considered. In this study, first the thermodynamic parameters, i.e. solubility product and the reaction enthalpy change of struvite precipitation under marine/brackish aquaculture conditions, were estimated based on experimental results from artificial marine/ brackish water and filtrate of digestate from an anaerobic digester treating sludge from a brackish RAS. Experimental results were compared with values reported in literature. In addition, pH adjustments of the digestate from the digester were carried out to investigate phosphate solubilization and precipitation. More-
over, the effects of pH and temperature on particle size distribution (PSD) of the formed precipitates in artificial brackish water were investigated. The PSD of struvite formed in the synthetic brackish water was compared with the PSD of struvite formed in the actual filtrate. The latter elucidates the impact of the water matrix on struvite formation under the marine/brackish conditions. 2. Material and method 2.1. Reagent, stock solution and digestate The artificial brackish water for tests of struvite precipitation (No. 1–13 and 18–21) was prepared as listed in Table 1. All the used chemicals purchased from Sigma (The Netherlands) were analytical grade. The digestate was collected from a lab-scale anaerobic digester (4 L) fed with sludge from a brackish aquaculture recirculation system. 2.2. Experimental set-up for struvite precipitation tests and pH adjustment of digestate Tests of struvite precipitation were carried out in 1 L beakers with a jar test apparatus (VELP-JLT6). 500 mL of the synthetic marine/brackish solution (Table 1) or filtrate of the digestate from an anaerobic digester treating sludge from a brackish RAS, supplemented with varying amounts of MgCl2 stock solution, were mixed in the beaker. During the test, the beaker was covered with a piece of parafilm to minimize NH3 volatilization. Preliminary struvite precipitation tests were carried out to assess the dynamics of ammonia and reactive phosphorus concentrations. The results (Fig. 1) show that the equilibria of ammonia and reactive phosphorus were reached within 75 min. Thus, in the study, 75 min were allowed for each test to ensure that equilibrium was reached. Solutions were stirred with paddles at 300 rpm over 75 min. At the end of the experiments, samples were taken to analyze the particle size distribution. Then, a 10 mL sample was filtered with membrane filters (0.45 lm, Whatman) and afterward stored in a fridge at 4 °C for later ion analyses. Digestate with a pH of 7.7 from the lab-scale anaerobic digester treating sludge from a brackish RAS was collected. pH adjustments of the digestate using 10 M HCl and 12.5 M NaOH solutions were
Table 1 Initial values of the tests of artificial brackish waters and the filtrates. Test No.
Temperature (°C)
Ionic strength (M)
[NH4] (mM)
[Portho] (mM)
Mg/P initial
K (mM)
Na (mM)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
25 25 25 25 25 25 25 25 25 4 32 46 56 25 25 25 25 25 25 25 25
0.50 0.59 0.67 0.53 0.44 0.43 0.41 0.38 0.37 0.43 0.43 0.43 0.43
243 286 357 257 179 179 179 150 143 179 179 179 179 226 230 220 227 179 179 179 179
16 20 19 14 12 10 8 6 5 10 10 10 10 12 23 5 6 10 10 10 10
0.12 0.41 0.58 0.78 0.91 1.09 1.37 1.82 2.19 1.09 1.09 1.09 1.09 1.00 0.50 1.44 2.19 1.09 1.09 1.09 1.09
5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 19.2 18.9 19.2 18.9 5.12 5.12 5.12 5.12
198 210 214 192 196 180 174 175 172 180 180 180 180 280 262 250 245 180 180 180 180
–
– – – 0.43 0.43 0.43 0.43
Note: –, not available; No. 1 to 13 were synthetic brackish waters; No. 14 to 17 were filtrate that was harvested from the digestate of the digester using a membrane filter with a pore size of 0.45 lm (Whatman, Germany). Assays No. 18 to No. 21 were the tests for different pH values using the same recipe with No. 6.
X. Zhang et al. / Bioresource Technology 218 (2016) 1151–1156
1153
attributed to the negligence of the effect of ionic strength on the product. Therefore, for calculation of the struvite solubility product under marine/brackish conditions, the ionic strength shall be considered. I is defined as shown in Eq. (7.3) (Stumm and Morgan, 1996).
I ¼ 0:5
X ðC i z2i Þ
ð7:3Þ
i
where ci is the concentration of ion i and zi the valence of ion i. Struvite formation is also temperature dependent. The standard enthalpy change of this reaction can be estimated via Van’t Hoff equation as shown in Eq. (7.4).
d ln K sp DHhr ¼ dT RT 2 Fig. 1. Reactive phosphorus (PO4-P) and ammonia-nitrogen (NH4-N) concentrations during two tests over time, T1 and T2 represent the two preliminary tests.
carried out to examine the solubilization and precipitation of struvite under the marine/brackish conditions.
Ammonia was measured immediately after sampling using an ammonia assay kit (Merck, Germany) and the other ions such as magnesium and phosphate were measured using a spectro arcosEOP spectrometer (Spectro Analytical Instruments, Germany). The PSD of precipitates was measured using HIAC Particle Counter-Model 3000 (USA), which covers a range from 2 to 400 lm. All PSD measurements of the samples were conducted in a series of dilution using its individual filtrate from each test to maintain the equilibrium condition. The dilution factors were as follows: 20, 40, 80, 100, and 200. Then the average of the volumetric percentages at the corresponding diameter of precipitates was used to present the PSD of the test. Precipitate samples were dried in an oven at 103–105 °C. Then purity of the dried struvite sample was analyzed by X-ray Diffraction (XRD) analysis using a Bruker D5005 diffractometer equipped with Huber incidentbeam monochromator. 2.4. Definition of thermodynamic solubility product of struvite and parameter estimation of the standard solubility product and standard enthalpy change of struvite formation The thermodynamic standard solubility product of struvite is defined as in Eq. (7.1).
K osp
2þ
¼ fMg g
fNHþ4 g
fPO3 4 g
K osp
2þ
or fMg g ¼
fNHþ4 g fPO3 4 g
ð7:1Þ
which can also be written by using the respective activity coefficients (Eq. (7.2)),
K osp
where T presents temperature in Kelvin and R is the gas constant (8.314 J K1 mol1). Because the standard enthalpy (DH0r ) varies slightly over temperature and can be considered constant for the range of interest (Bhuiyan et al., 2007), therefore Eq. (7.4) can be integrated and rewritten as Eq. (7.5).
ln K sp ¼
2.3. Analytical methods
¼ cMg2þ ½Mg cNHþ 2þ
4
½NHþ4
cPO3 4
½PO3 4
ð7:2Þ
where K osp represents the thermodynamic solubility product, {} represents the activity of the ion, [] indicates the concentration of the ion, and c represents the activity coefficient of the corresponding ion. Hereby, the conditional solubility product of struvite is defined 3 as the product of ½Mg 2þ , ½NHþ 4 and ½PO4 . It is generally considered that the activity coefficient is a function of the ionic strength of the solution (Bhuiyan et al., 2007). Hanhoun et al. (2011) and Bhuiyan et al. (2007) indicated that the discrepancies between the reported values of the thermodynamic solubility products of struvite in literature were partially
ð7:4Þ
DH0r 1 þC R T
ð7:5Þ
where C is a constant. The chemical speciation program PHREEQC is often used by researchers because of its well-established database on thermodynamic data and robust numerical engine. In the current study, PHREEQC (version 2.18) was employed to calculate the equilibrium concentrations of ions. The activity coefficients of ions were calculated using the Davies approximation shown in Eq. (7.6) (Ronteltap et al., 2007). It has been reported that the Davies approximation shows the best agreement between the calculated results and experimental data in struvite precipitation tests, using urine characterized by high ionic strength (Ronteltap et al., 2007).
log ci ¼
Az2i
! pffiffi I pffiffi BI 1þ I
ð7:6Þ
where ci is the activity coefficient of ion i, and A and B are constants. All the equilibrium concentrations and speciation of ions measured in the experimental tests were calculated by using PHREEQC 2.18. Using Eqs. (7.1) and (7.5), the thermodynamic solubility product and the standard enthalpy change of struvite formation were estimated by nonlinear parameter estimation fitting using Matlab, respectively. 3. Results and discussion 3.1. Struvite crystallization in digestate from an anaerobic reactor treating sludge from a brackish recirculation aquaculture system At pH levels of 2.0, 7.7, and 9.0, reactive phosphorus (RP) concentrations of the digestates in batch assays that were sampled from an anaerobic reactor fed with sludge from the brackish RAS were 1288, 46, and 34 mg PO4-P/L, respectively, as also depicted in Fig. 2. With the further increase in pH from 10 to 12, reactive phosphorus increased to 247 mg PO4-P/L (Fig. 2). The trend was also observed in the study conducted by Nelson et al. (2003). The results indicate that in the digester a large fraction of phosphorus that was released from bound phosphorus during AD was precipitated at pH 7.7, i.e. the natural pH value of the digester. This can be explained by precipitation of struvite. In the influent of the digester, the magnesium concentration was 347 mg/L originating from the brackish RAS water. With the high amount of ammonia produced during AD, struvite precipitation, therefore, may be the
1154
X. Zhang et al. / Bioresource Technology 218 (2016) 1151–1156
Fig. 2. NH4-N, Mg2+ and PO4-P in digestate at different pH levels.
dominant form of phosphorus precipitation in the digester. Moreover, the results show that the existing composition of the digestate already offered the proper condition for struvite precipitation in the digester, where only 3.6% of reactive phosphorus was remaining, compared to the reactive phosphorus of the digestate (1288 mg PO4-P/L) when its pH was adjusted to 2.0. The products of the concentrations of Mg2+, NH+4 and PO3 at 4 different pH levels (Fig. S1) show that when pH increased above 5, the product declined sharply. This was probably due to the increased fraction of PO3 in the total reactive phosphorus at 4 PO3 4 ,
possibly resulting in higher pH, that is, increased activity of enhancement of struvite formation. With the increase in pH from 10, the product seems to start increasing. The results (Fig. S1) also suggest that there was a minimum of the product at a pH around 10. With the increased pH from 10, reactive phosphorus increased, which may be due to solubilization of struvite. It is because the further increase in pH led to the increased free ammonia and ammonia gas fractions, also formation of Mg complex such as (MgOH)+, and the precipitation of Mg salts such as Mg(OH)2, i.e. decreases in activities of Mg2+ and NH+4. The results (Fig. S1) may indicate that struvite has a minimal thermodynamic solubility product at a pH around 10. Nelson et al. (2003) reported a minimal solubility of struvite under fresh water conditions in the pH ranging from 8.9 to 9.3. The difference could be related to the disparity of ionic strength in the present study and the study conducted by Nelson et al. (2003).
3.2. Solubility product and standard enthalpy of struvite formation under brackish condition (25 °C) All the equilibrium activities of Mg2+, NH+4 and PO3 of the 4 assays (No.1–9 and 14–17) listed in Table 1 were calculated using PHREEQC 2.18. Afterward, based on the activities and Eq. (7.2), nonlinear regression was conducted between the activities to estimate the thermodynamic solubility product K osp (25 °C). The esti-
was 25.7 kJmol1 in this study. However, in two other studies enthalpy changes of struvite formation of 22.6 kJ mol1 (Ronteltap et al., 2007) and 24.1 kJ mol1 (Bhuiyan et al., 2007) were reported. The relatively larger enthalpy change observed in this study was probably due to the larger temperature range (4– 56 °C) for the assays, as Bhuiyan et al. (2007) stated that using the same enthalpy for struvite formation for the whole temperature range could lead to erroneous results. In addition, the disparity in enthalpy change in this present study and Bhuiyan et al. (2007) could also be related to the differences in ionic strength. That is, ionic strength ranging from 0.01to 0.1 M in Bhuiyan et al. (2007) and from 0.37 to 0.67 M (Table 1) in this current study. However, the difference between the estimated enthalpy change in this study and the ones published by Bhuiyan et al. (2007) and Ronteltap et al. (2007) is not substantial, which may also indicate that even in the larger temperature range, the influence of temperature variation on the enthalpy change of struvite formation may still be limited. Thus, the assumption, considering the enthalpy change of struvite formation as constant in the whole temperature range, is still valid. Therefore, the estimated K osp (1013.06) and DH0r (25.7 kJ mol1) of struvite obtained in this study can be used to predict struvite precipitation in marine/brackish aquaculture systems and anaerobic reactors treating sludges from the marine/ brackish aquaculture systems.
3.3. Effect of pH on struvite particle size distribution under the marine/ brackish condition The composition of ions and particle size distribution of precipitates in the four batch tests at pH 6.7, 8.3, 9.0, and 9.7 (No. 18 to 21) were analyzed at the end of the tests. Fig. 3 shows that with the increase in pH, the crystal size of struvite decreases in the pH range. Matynia et al. (2006) also reported that an increase in pH from 8 to 11 decreased the mean crystal size of struvite formed in synthetic solutions with a factor 5.5 times. This observed phenomenon was likely related to the supersaturation of struvite (Ronteltap et al., 2010). The supersaturation indices of the four tests (No. 18 to No. 21) were calculated using PHREEQC, as follows 1.33, 2.38, 2.80, and 3.01 at a pH of 6.7, 8.3, 9.0, and 9.7 respectively. At high supersaturation index, struvite nucleation outweighs crystal growth, and therefore a high fraction of struvite appeared as crystals with small size or even undetectable nuclei. Moreover, based on the report of (Le Corre et al., 2009), the size of struvite crystals may be restricted by electrostatic repulsion. Individual struvite particles usually present a negative zeta-potential, which limits agglomeration of particles. As pH increases its zeta-potential becomes further negative, possibly leading to the decreases in crystal size.
mated K osp of struvite under the marine/brackish condition was 1013.06 with an estimated standard deviation of 100.08. Ronteltap et al. (2007) reported that the thermodynamic solubility product of struvite in hydrolyzed urine is 1013.26, which is rather close to the value estimated in this study. Based on the results of the assays (No. 6 and No. 10–13) and using Eq. (7.5), nonlinear regression was employed to estimate the standard enthalpy change of struvite formation DHhr with the assumption that DHhr only varies slightly with temperature and could be regarded as constant with temperature within a range of few tens of degrees (Bhuiyan et al., 2007). The estimated DHhr
Fig. 3. Particle size distribution of struvite of the tests with various pHs.
X. Zhang et al. / Bioresource Technology 218 (2016) 1151–1156
Moreover, among the four tests of particle size distribution, the groups with pH values of 6.7 and 8.3 yielded double peaks at 50 lm and 90 lm for the test at pH 6.7, and 40 lm and 70 lm for the test at pH of 8.3. These results indicate a possible change in crystal growth kinetics during struvite crystallization. This might be related to a variation of the surface charge, or surface influence, from the increased amounts of H2PO 4 at a pH of around 7, which was also reported in the study (Ronteltap et al., 2010). Moreover, XRD results (Fig. S2) show larger presence intensity of newberyite (HMgPO43H2O) in the assay at a pH of 8.3 than in the assay of a pH 9.0. Thus, the splits of peaks at pH values of 6.7 and 8.3 might be related to the formation of newberyite because the formation of struvite and newberyite is also as function of pH (Abbona et al., 1982).
3.4. Effect of temperature on solubility product and particle size distribution of struvite under the marine/brackish condition The equilibrium activities of Mg2+, NH+4 and PO3 4 of the assays (No. 10–13) were calculated using PHREEQC. The trend of ion activity product (IAP), i.e. product of activities of Mg2+, NH+4 and PO3 4 , as function of temperature shown in Fig. S3 agreed well with the results reported by Aage et al. (1997) and Burns and Finlayson (1982). That is: the solubility product of struvite increases over temperature. In contrast, Bhuiyan et al. (2007) reported that struvite solubility product increased with temperature in the range of 10 to 35 °C and decreased in the range from 35 to 60 °C. They hypothesized that this was due to the transition of struvite, that is, from struvite (NH4MgPO46H2O) to dittmarite (NH4MgPO4H2O). However, the XRD results of crystals formed at a temperature of 46 °C in No. assay 12 and at a temperature of 56 °C in No. 13 in this present study show that the crystals were NH4MgPO46H2O and newberyite, and no dittmarite was observed, possibly explaining the difference between the results in this current study and those of Bhuiyan et al. (2007). Fig. 4 shows that crystal size increased with the increase in temperature, and also the distribution seems to be relatively more even at 46 and 56 °C than at 4 °C. Temperature substantially influences crystal growth rate (Bhuiyan et al., 2008). At higher temperature, struvite has higher solubility, and the solution became less supersaturation, which is favorable for struvite crystal growth, rather than for nucleation (Desmidt et al., 2013; Ronteltap et al., 2007). In addition, at 46 and 56 °C, two peaks in the PSD were observed, which was also observed in the study of Ronteltap et al. (2010). This may be related to the co-existence of struvite and newberyite as the XRD results shown in Fig. S2.
1155
3.5. Comparison of composition of crystals formed in synthetic brackish water and digestate filtrate XRD analyses (Fig. S4) show that the dominant crystals formed from digestate filtrate consisted of N-struvite that contains ammonia and dittmarite, whereas the crystals of synthetic brackish water consisted of struvite and newberyite (Fig. S2). That difference in composition is likely related to the difference in the initial NH4-N concentrations, that is: 179 mM in the synthetic brackish water test No. 6 and 220 mM in the filtrate. The final equilibrium NH4-N concentrations in the synthetic brackish water and the filtrate were 177 mM and 208 mM, respectively. The higher concentration of ammonia likely enhanced the formation of ammonia compounds such as struvite and dittmarite instead of the amorphous magnesium and calcium phosphates (Crutchik and Garrido, 2011; Crutchik et al., 2013). In addition, in this study Kstruvite that contains potassium instead of ammonia was not observed based on the XRD results probably due to the relative low concentration of potassium ranging from 5.12 to 19.2 mM, compared to the concentration of ammonia (Table 1). It is pronounced that the characteristics of (waste)waters played also an important role on the nature of precipitates containing ammonia and phosphate, which was also observed by Crutchik et al. (2013).
4. Conclusions In this study, it can be concluded that: 1) In the digester, struvite was formed due to the naturally present magnesium and the released ammonia and phosphate, which kept phosphate concentration low in the digester; the digester with pH of 7.7 delivered proper condition for struvite formation. 2) Estimated thermodynamic solubility product (K osp ) and estimated enthalpy change (DHhr ) of struvite formation under and the marine/brackish condition are 1013.06 o 1 25.7 kJ mol , respectively. The estimated K sp and DHhr coincide with the literature values obtained under high salinity conditions. The estimated parameters of struvite formation can be used to predict struvite precipitation in marine/ brackish aquaculture systems and anaerobic reactors treating the salty RAS sludges. 3) The average crystal size of precipitates under the marine/ brackish condition decreased with pH. 4) The average crystal size increased with temperature under the marine/brackish condition. 5) High NH+4 enhanced formation of the ammonia precipitates struvite and dittmarite.
Acknowledgement The authors acknowledge EM-MARES project and Agentschap NL for financial support to this research and China Scholarship Council for granting scholarships to Xuedong Zhang.
Appendix A. Supplementary data Fig. 4. The particle size distribution of the tests No. 6, 10, 11, 12 and 13 with various temperatures.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2016.07. 088.
1156
X. Zhang et al. / Bioresource Technology 218 (2016) 1151–1156
References Aage, H.K., Andersen, B.L., Blom, A., Jensen, I., 1997. The solubility of struvite. J. Radioanal Nucl. Chem. 223 (1–2), 213–215. Abbona, F., Lundager Madsen, H.E., Boistelle, R., 1982. Crystallization of two magnesium phosphates, struvite and newberyite: effect of pH and concentration. J. Cryst. Growth 57 (1), 6–14. Bhuiyan, M.I.H., Mavinic, D.S., Beckie, R.D., 2008. Nucleation and growth kinetics of struvite in a fluidized bed reactor. J. Cryst. Growth 310 (6), 1187–1194. Bhuiyan, M.I.H., Mavinic, D.S., Beckie, R.D., 2007. A solubility and thermodynamic study of struvite. Environ. Technol. 28 (9), 1015–1026. Burns, J.R., Finlayson, B., 1982. Solubility product of magnesium ammonium phosphate hexahydrate at various temperatures. J. Urol. 128 (2), 426–428. Cordell, D., Drangert, J.O., White, S., 2009. The story of phosphorus: global food security and food for thought. Global Environ. Change 19 (2), 292–305. Crutchik, D., Garrido, J.M., 2011. Struvite crystallization versus amorphous magnesium and calcium phosphate precipitation during the treatment of a saline industrial wastewater. Water Sci. Technol. 64 (12), 2460–2467. Crutchik, D., Sanchez, A., Garrido, J.M., 2013. Simulation and experimental validation of multiple phosphate precipitates in a saline industrial wastewater. Sep. Purif. Technol. 118, 81–88. Desmidt, E., Ghyselbrecht, K., Monballiu, A., Rabaey, K., Verstraete, W., Meesschaert, B.D., 2013. Factors influencing urease driven struvite precipitation. Sep. Purif. Technol. 110, 150–157. Hanhoun, M., Montastruc, L., Azzaro-Pantel, C., Biscans, B., Freche, M., Pibouleau, L., 2011. Temperature impact assessment on struvite solubility product: a thermodynamic modeling approach. Chem. Eng. J. 167 (1), 50–58. He, Z.W., Liu, W.Z., Wang, L., Yang, C.X., Guo, Z.C., Zhou, A.J., Liu, J.Y., Wang, A.J., 2016. Role of extracellular polymeric substances in enhancement of phosphorus release from waste activated sludge by rhamnolipid addition. Bioresour. Technol. 202, 59–66. Lahav, O., Telzhensky, M., Zewuhn, A., Gendel, Y., Gerth, J., Calmano, W., Birnhack, L., 2013. Struvite recovery from municipal-wastewater sludge centrifuge
supernatant using seawater NF concentrate as a cheap Mg(II) source. Sep. Purif. Technol. 108, 103–110. Le Corre, K.S., Valsami-Jones, E., Hobbs, P., Parsons, S.A., 2009. Phosphorus recovery from wastewater by struvite crystallization: a review. Crit. Rev. Environ. Sci. Technol. 39 (6), 433–477. Liu, B.X., Giannis, A., Zhang, J.F., Chang, V.W.C., Wang, J.Y., 2013. Characterization of induced struvite formation from source-separated urine using seawater and brine as magnesium sources. Chemosphere 93 (11), 2738–2747. Matynia, A., Koralewska, J., Wierzbowska, B., Piotrowski, K., 2006. The influence of process parameters on struvite continuous crystallization kinetics. Chem. Eng. Commun. 193 (2), 160–176. Nelson, N.O., Mikkelsen, R.L., Hesterberg, D.L., 2003. Struvite precipitation in anaerobic swine lagoon liquid: effect of pH and Mg: P ratio and determination of rate constant. Bioresour. Technol. 89 (3), 229–236. O’Neal, J.A., Boyer, T.H., 2013. Phosphate recovery using hybrid anion exchange: applications to source-separated urine and combined wastewater streams. Water Res. 47 (14), 5003–5017. Ronteltap, M., Maurer, M., Gujer, W., 2007. Struvite precipitation thermodynamics in source-separated urine. Water Res. 41 (5), 977–984. Ronteltap, M., Maurer, M., Hausherr, R., Gujer, W., 2010. Struvite precipitation from urine – influencing factors on particle size. Water Res. 44 (6), 2038–2046. Stumm, W., Morgan, J.J., 1996. Aquatic Chemistry: Chemical Equalibria and Rates in Natural Waers, third ed. Wiley Interscience, New York. Vaccari, D.A., 2009. Phosphorus: a looming crisis. Sci. Am. 300 (6), 54–59. Xu, Y.F., Hu, H., Liu, J.Y., Luo, J.H., Qian, G.R., Wang, A.J., 2015. PH dependent phosphorus release from waste activated sludge: contributions of phosphorus speciation. Chem. Eng. J. 267, 260–265. Zhang, X., Spanjers, H., van Lier, J.B., 2013. Potentials and limitations of biomethane and phosphorus recovery from sludges of brackish/marine aquaculture recirculation systems: a review. J. Environ. Manage. 131, 44–54. Zhang, X., Tao, Y., Hu, J., Liu, G., Spanjers, H., van Lier, J.B., 2016. Biomethanation and microbial community changes in a digester treating sludge from a brackish aquaculture recirculation system. Bioresour. Technol. 214, 338–347.
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Struvite crystallization under a marine/brackish aquaculture condition Xuedong Zhang ⇑, Jianmei Hu, Henri Spanjers, Jules B. van Lier Section Sanitary Engineering, Department of Water Management, Delft University of Technology, Stevinweg 1, 2628CN Delft, The Netherlands
h i g h l i g h t s ) and DH0r (25.7 kJ mol1) of struvite under the saline condition were obtained. 2+ and release of NH4+ led to low PO3 4 in the digester. Minimal solubility of struvite under marine/brackish conditions was at pH around 10. Average crystal size of struvite under the conditions decreased with pH. + High NH4 enhanced formation of ammonia precipitates, e.g. struvite and dittmarite. o
K sp (10
13.06
Natural presence of Mg
a r t i c l e
i n f o
Article history: Received 25 May 2016 Received in revised form 19 July 2016 Accepted 20 July 2016 Available online 21 July 2016 Keywords: Struvite crystallization Marine/brackish aquaculture condition Particle size distribution Thermodynamic solubility product Standard enthalpy
a b s t r a c t The results in this study show that struvite was formed in the digester at pH 7.7 due to the magnesium naturally present and the released ammonia and phosphate, resulting in low phosphate concentration in the digester. Apparently the digester already provided proper conditions for struvite formation. Under the brackish condition, the estimated thermodynamic solubility product and enthalpy change of struvite formation were 1013.06 and 25.7 kJ mol1, respectively. The average crystal size under marine/brackish condition decreased with pH, but increased with temperature. X-ray diffraction measurements indicate struvite (NH4MgPO46H2O) and dittmarite (NH4MgPO4H2O) were predominant phosphorus species produced in filtrates of the digester. However, struvite and newberyite (HMgPO43H2O) were the predominant species precipitated from synthetic brackish waters after dosing MgCl2. It is pronounced that (waste)water characteristics played also an important role on the nature of phosphate precipitates. Under high NH+4 condition, phosphorus precipitates containing ammonia were dominant, compared to other amorphous phosphates. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Nowadays phosphorus ending up in waste streams such as effluents from recirculation aquaculture systems (RAS) is a concern because the emission of phosphorus causes eutrophication of local water bodies. Meanwhile, steady supply of P fertilizer is essential for sustainable agricultural production to feed the soaring population (Cordell et al., 2009). In recent years there is an increasing concern about the exhaustion of exploitable phosphorus reserves (Vaccari, 2009; Xu et al., 2015), leading to a decline in phosphate rock production in some countries and finally resulting in the soaring price of phosphate rock. Due to the aforementioned issues, there is an increasing attention to investigation of struvite (NH4MgPO46H2O) precipitation and thereby recovery of phosphorus from various types of P rich waste streams, such as urine and ⇑ Corresponding author. E-mail addresses: [email protected], [email protected] (X. Zhang). http://dx.doi.org/10.1016/j.biortech.2016.07.088 0960-8524/Ó 2016 Elsevier Ltd. All rights reserved.
rejected water from digesters (He et al., 2016; Lahav et al., 2013; Liu et al., 2013; O’Neal and Boyer, 2013; Ronteltap et al., 2007, 2010). However, to date no report on struvite crystallization/precipitation under an aquaculture brackish condition is available yet, although there is a potential to recover P from P-rich sludge from aquaculture recirculation systems. Most studies on struvite crystallization/precipitation are carried out under low ionic strength conditions. Several studies on struvite precipitation under high ionic strength are available (Bhuiyan et al., 2007; Crutchik and Garrido, 2011; Crutchik et al., 2013; Liu et al., 2013; O’Neal and Boyer, 2013; Ronteltap et al., 2007, 2010). The ionic strength influences activity coefficients of ions, particularly the multivalent ions, such as Mg2+ and PO3 4 , which may further affect the thermodynamic parameters of struvite formation, such as the thermodynamic solubility product and the enthalpy of struvite formation. Theoretically, under high ionic strength conditions, struvite solubility increases due to the decreased activity coefficients of Mg2+, NH+4 and PO3 4 . However,
1152
X. Zhang et al. / Bioresource Technology 218 (2016) 1151–1156
limited literature on struvite behavior under high ionic strength conditions, i.e. marine/brackish conditions, is available, and thus further investigation on struvite formation under the high salinity condition is needed. In marine/brackish RAS, spontaneous struvite formation could occur, due to the naturally present Mg in the marine/brackish culturing water, and ammonia and phosphate released from degradation of fish faeces and/or fish feed (Zhang et al., 2013, 2016). Therefore, uncontrolled struvite scaling might become a problem for the piping system of marine/brackish RAS. Moreover, when anaerobic digestion for sludge stabilization is applied, the production of a P- and N- rich stream in and from the digestate will further enhance struvite precipitation. Furthermore, the digestate from a digester with controlled struvite precipitation in treating the salty sludge could further serve as fertilizer for salty plants. Therefore, a better understanding of struvite precipitation under marine/brackish aquaculture conditions is conducive to avoiding struvite scaling and to better unveiling struvite precipitation in the anaerobic digestion process treating saline (waste)waters. It is commonly regarded that pH and supersaturation have major impacts on struvite precipitation (Hanhoun et al., 2011). In addition, temperature as well affects struvite solubility and the enthalpy change of struvite formation (Bhuiyan et al., 2007; Hanhoun et al., 2011). Thus, the effects of pH, supersaturation, and temperature on struvite precipitation under marine/brackish aquaculture conditions are of great interest to be investigated. This study aims to understand struvite precipitation in digestates from an anaerobic digester treating the salty sludge from a marine/brackish RAS. Thus, in order to simplify the analyses in the batch tests, artificial mimicked supernatants of digestate were employed, and precipitation of the precipitates, such as calcium phosphate, calcium carbonate, and magnesium carbonate, was not considered. In this study, first the thermodynamic parameters, i.e. solubility product and the reaction enthalpy change of struvite precipitation under marine/brackish aquaculture conditions, were estimated based on experimental results from artificial marine/ brackish water and filtrate of digestate from an anaerobic digester treating sludge from a brackish RAS. Experimental results were compared with values reported in literature. In addition, pH adjustments of the digestate from the digester were carried out to investigate phosphate solubilization and precipitation. More-
over, the effects of pH and temperature on particle size distribution (PSD) of the formed precipitates in artificial brackish water were investigated. The PSD of struvite formed in the synthetic brackish water was compared with the PSD of struvite formed in the actual filtrate. The latter elucidates the impact of the water matrix on struvite formation under the marine/brackish conditions. 2. Material and method 2.1. Reagent, stock solution and digestate The artificial brackish water for tests of struvite precipitation (No. 1–13 and 18–21) was prepared as listed in Table 1. All the used chemicals purchased from Sigma (The Netherlands) were analytical grade. The digestate was collected from a lab-scale anaerobic digester (4 L) fed with sludge from a brackish aquaculture recirculation system. 2.2. Experimental set-up for struvite precipitation tests and pH adjustment of digestate Tests of struvite precipitation were carried out in 1 L beakers with a jar test apparatus (VELP-JLT6). 500 mL of the synthetic marine/brackish solution (Table 1) or filtrate of the digestate from an anaerobic digester treating sludge from a brackish RAS, supplemented with varying amounts of MgCl2 stock solution, were mixed in the beaker. During the test, the beaker was covered with a piece of parafilm to minimize NH3 volatilization. Preliminary struvite precipitation tests were carried out to assess the dynamics of ammonia and reactive phosphorus concentrations. The results (Fig. 1) show that the equilibria of ammonia and reactive phosphorus were reached within 75 min. Thus, in the study, 75 min were allowed for each test to ensure that equilibrium was reached. Solutions were stirred with paddles at 300 rpm over 75 min. At the end of the experiments, samples were taken to analyze the particle size distribution. Then, a 10 mL sample was filtered with membrane filters (0.45 lm, Whatman) and afterward stored in a fridge at 4 °C for later ion analyses. Digestate with a pH of 7.7 from the lab-scale anaerobic digester treating sludge from a brackish RAS was collected. pH adjustments of the digestate using 10 M HCl and 12.5 M NaOH solutions were
Table 1 Initial values of the tests of artificial brackish waters and the filtrates. Test No.
Temperature (°C)
Ionic strength (M)
[NH4] (mM)
[Portho] (mM)
Mg/P initial
K (mM)
Na (mM)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
25 25 25 25 25 25 25 25 25 4 32 46 56 25 25 25 25 25 25 25 25
0.50 0.59 0.67 0.53 0.44 0.43 0.41 0.38 0.37 0.43 0.43 0.43 0.43
243 286 357 257 179 179 179 150 143 179 179 179 179 226 230 220 227 179 179 179 179
16 20 19 14 12 10 8 6 5 10 10 10 10 12 23 5 6 10 10 10 10
0.12 0.41 0.58 0.78 0.91 1.09 1.37 1.82 2.19 1.09 1.09 1.09 1.09 1.00 0.50 1.44 2.19 1.09 1.09 1.09 1.09
5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 5.12 19.2 18.9 19.2 18.9 5.12 5.12 5.12 5.12
198 210 214 192 196 180 174 175 172 180 180 180 180 280 262 250 245 180 180 180 180
–
– – – 0.43 0.43 0.43 0.43
Note: –, not available; No. 1 to 13 were synthetic brackish waters; No. 14 to 17 were filtrate that was harvested from the digestate of the digester using a membrane filter with a pore size of 0.45 lm (Whatman, Germany). Assays No. 18 to No. 21 were the tests for different pH values using the same recipe with No. 6.
X. Zhang et al. / Bioresource Technology 218 (2016) 1151–1156
1153
attributed to the negligence of the effect of ionic strength on the product. Therefore, for calculation of the struvite solubility product under marine/brackish conditions, the ionic strength shall be considered. I is defined as shown in Eq. (7.3) (Stumm and Morgan, 1996).
I ¼ 0:5
X ðC i z2i Þ
ð7:3Þ
i
where ci is the concentration of ion i and zi the valence of ion i. Struvite formation is also temperature dependent. The standard enthalpy change of this reaction can be estimated via Van’t Hoff equation as shown in Eq. (7.4).
d ln K sp DHhr ¼ dT RT 2 Fig. 1. Reactive phosphorus (PO4-P) and ammonia-nitrogen (NH4-N) concentrations during two tests over time, T1 and T2 represent the two preliminary tests.
carried out to examine the solubilization and precipitation of struvite under the marine/brackish conditions.
Ammonia was measured immediately after sampling using an ammonia assay kit (Merck, Germany) and the other ions such as magnesium and phosphate were measured using a spectro arcosEOP spectrometer (Spectro Analytical Instruments, Germany). The PSD of precipitates was measured using HIAC Particle Counter-Model 3000 (USA), which covers a range from 2 to 400 lm. All PSD measurements of the samples were conducted in a series of dilution using its individual filtrate from each test to maintain the equilibrium condition. The dilution factors were as follows: 20, 40, 80, 100, and 200. Then the average of the volumetric percentages at the corresponding diameter of precipitates was used to present the PSD of the test. Precipitate samples were dried in an oven at 103–105 °C. Then purity of the dried struvite sample was analyzed by X-ray Diffraction (XRD) analysis using a Bruker D5005 diffractometer equipped with Huber incidentbeam monochromator. 2.4. Definition of thermodynamic solubility product of struvite and parameter estimation of the standard solubility product and standard enthalpy change of struvite formation The thermodynamic standard solubility product of struvite is defined as in Eq. (7.1).
K osp
2þ
¼ fMg g
fNHþ4 g
fPO3 4 g
K osp
2þ
or fMg g ¼
fNHþ4 g fPO3 4 g
ð7:1Þ
which can also be written by using the respective activity coefficients (Eq. (7.2)),
K osp
where T presents temperature in Kelvin and R is the gas constant (8.314 J K1 mol1). Because the standard enthalpy (DH0r ) varies slightly over temperature and can be considered constant for the range of interest (Bhuiyan et al., 2007), therefore Eq. (7.4) can be integrated and rewritten as Eq. (7.5).
ln K sp ¼
2.3. Analytical methods
¼ cMg2þ ½Mg cNHþ 2þ
4
½NHþ4
cPO3 4
½PO3 4
ð7:2Þ
where K osp represents the thermodynamic solubility product, {} represents the activity of the ion, [] indicates the concentration of the ion, and c represents the activity coefficient of the corresponding ion. Hereby, the conditional solubility product of struvite is defined 3 as the product of ½Mg 2þ , ½NHþ 4 and ½PO4 . It is generally considered that the activity coefficient is a function of the ionic strength of the solution (Bhuiyan et al., 2007). Hanhoun et al. (2011) and Bhuiyan et al. (2007) indicated that the discrepancies between the reported values of the thermodynamic solubility products of struvite in literature were partially
ð7:4Þ
DH0r 1 þC R T
ð7:5Þ
where C is a constant. The chemical speciation program PHREEQC is often used by researchers because of its well-established database on thermodynamic data and robust numerical engine. In the current study, PHREEQC (version 2.18) was employed to calculate the equilibrium concentrations of ions. The activity coefficients of ions were calculated using the Davies approximation shown in Eq. (7.6) (Ronteltap et al., 2007). It has been reported that the Davies approximation shows the best agreement between the calculated results and experimental data in struvite precipitation tests, using urine characterized by high ionic strength (Ronteltap et al., 2007).
log ci ¼
Az2i
! pffiffi I pffiffi BI 1þ I
ð7:6Þ
where ci is the activity coefficient of ion i, and A and B are constants. All the equilibrium concentrations and speciation of ions measured in the experimental tests were calculated by using PHREEQC 2.18. Using Eqs. (7.1) and (7.5), the thermodynamic solubility product and the standard enthalpy change of struvite formation were estimated by nonlinear parameter estimation fitting using Matlab, respectively. 3. Results and discussion 3.1. Struvite crystallization in digestate from an anaerobic reactor treating sludge from a brackish recirculation aquaculture system At pH levels of 2.0, 7.7, and 9.0, reactive phosphorus (RP) concentrations of the digestates in batch assays that were sampled from an anaerobic reactor fed with sludge from the brackish RAS were 1288, 46, and 34 mg PO4-P/L, respectively, as also depicted in Fig. 2. With the further increase in pH from 10 to 12, reactive phosphorus increased to 247 mg PO4-P/L (Fig. 2). The trend was also observed in the study conducted by Nelson et al. (2003). The results indicate that in the digester a large fraction of phosphorus that was released from bound phosphorus during AD was precipitated at pH 7.7, i.e. the natural pH value of the digester. This can be explained by precipitation of struvite. In the influent of the digester, the magnesium concentration was 347 mg/L originating from the brackish RAS water. With the high amount of ammonia produced during AD, struvite precipitation, therefore, may be the
1154
X. Zhang et al. / Bioresource Technology 218 (2016) 1151–1156
Fig. 2. NH4-N, Mg2+ and PO4-P in digestate at different pH levels.
dominant form of phosphorus precipitation in the digester. Moreover, the results show that the existing composition of the digestate already offered the proper condition for struvite precipitation in the digester, where only 3.6% of reactive phosphorus was remaining, compared to the reactive phosphorus of the digestate (1288 mg PO4-P/L) when its pH was adjusted to 2.0. The products of the concentrations of Mg2+, NH+4 and PO3 at 4 different pH levels (Fig. S1) show that when pH increased above 5, the product declined sharply. This was probably due to the increased fraction of PO3 in the total reactive phosphorus at 4 PO3 4 ,
possibly resulting in higher pH, that is, increased activity of enhancement of struvite formation. With the increase in pH from 10, the product seems to start increasing. The results (Fig. S1) also suggest that there was a minimum of the product at a pH around 10. With the increased pH from 10, reactive phosphorus increased, which may be due to solubilization of struvite. It is because the further increase in pH led to the increased free ammonia and ammonia gas fractions, also formation of Mg complex such as (MgOH)+, and the precipitation of Mg salts such as Mg(OH)2, i.e. decreases in activities of Mg2+ and NH+4. The results (Fig. S1) may indicate that struvite has a minimal thermodynamic solubility product at a pH around 10. Nelson et al. (2003) reported a minimal solubility of struvite under fresh water conditions in the pH ranging from 8.9 to 9.3. The difference could be related to the disparity of ionic strength in the present study and the study conducted by Nelson et al. (2003).
3.2. Solubility product and standard enthalpy of struvite formation under brackish condition (25 °C) All the equilibrium activities of Mg2+, NH+4 and PO3 of the 4 assays (No.1–9 and 14–17) listed in Table 1 were calculated using PHREEQC 2.18. Afterward, based on the activities and Eq. (7.2), nonlinear regression was conducted between the activities to estimate the thermodynamic solubility product K osp (25 °C). The esti-
was 25.7 kJmol1 in this study. However, in two other studies enthalpy changes of struvite formation of 22.6 kJ mol1 (Ronteltap et al., 2007) and 24.1 kJ mol1 (Bhuiyan et al., 2007) were reported. The relatively larger enthalpy change observed in this study was probably due to the larger temperature range (4– 56 °C) for the assays, as Bhuiyan et al. (2007) stated that using the same enthalpy for struvite formation for the whole temperature range could lead to erroneous results. In addition, the disparity in enthalpy change in this present study and Bhuiyan et al. (2007) could also be related to the differences in ionic strength. That is, ionic strength ranging from 0.01to 0.1 M in Bhuiyan et al. (2007) and from 0.37 to 0.67 M (Table 1) in this current study. However, the difference between the estimated enthalpy change in this study and the ones published by Bhuiyan et al. (2007) and Ronteltap et al. (2007) is not substantial, which may also indicate that even in the larger temperature range, the influence of temperature variation on the enthalpy change of struvite formation may still be limited. Thus, the assumption, considering the enthalpy change of struvite formation as constant in the whole temperature range, is still valid. Therefore, the estimated K osp (1013.06) and DH0r (25.7 kJ mol1) of struvite obtained in this study can be used to predict struvite precipitation in marine/brackish aquaculture systems and anaerobic reactors treating sludges from the marine/ brackish aquaculture systems.
3.3. Effect of pH on struvite particle size distribution under the marine/ brackish condition The composition of ions and particle size distribution of precipitates in the four batch tests at pH 6.7, 8.3, 9.0, and 9.7 (No. 18 to 21) were analyzed at the end of the tests. Fig. 3 shows that with the increase in pH, the crystal size of struvite decreases in the pH range. Matynia et al. (2006) also reported that an increase in pH from 8 to 11 decreased the mean crystal size of struvite formed in synthetic solutions with a factor 5.5 times. This observed phenomenon was likely related to the supersaturation of struvite (Ronteltap et al., 2010). The supersaturation indices of the four tests (No. 18 to No. 21) were calculated using PHREEQC, as follows 1.33, 2.38, 2.80, and 3.01 at a pH of 6.7, 8.3, 9.0, and 9.7 respectively. At high supersaturation index, struvite nucleation outweighs crystal growth, and therefore a high fraction of struvite appeared as crystals with small size or even undetectable nuclei. Moreover, based on the report of (Le Corre et al., 2009), the size of struvite crystals may be restricted by electrostatic repulsion. Individual struvite particles usually present a negative zeta-potential, which limits agglomeration of particles. As pH increases its zeta-potential becomes further negative, possibly leading to the decreases in crystal size.
mated K osp of struvite under the marine/brackish condition was 1013.06 with an estimated standard deviation of 100.08. Ronteltap et al. (2007) reported that the thermodynamic solubility product of struvite in hydrolyzed urine is 1013.26, which is rather close to the value estimated in this study. Based on the results of the assays (No. 6 and No. 10–13) and using Eq. (7.5), nonlinear regression was employed to estimate the standard enthalpy change of struvite formation DHhr with the assumption that DHhr only varies slightly with temperature and could be regarded as constant with temperature within a range of few tens of degrees (Bhuiyan et al., 2007). The estimated DHhr
Fig. 3. Particle size distribution of struvite of the tests with various pHs.
X. Zhang et al. / Bioresource Technology 218 (2016) 1151–1156
Moreover, among the four tests of particle size distribution, the groups with pH values of 6.7 and 8.3 yielded double peaks at 50 lm and 90 lm for the test at pH 6.7, and 40 lm and 70 lm for the test at pH of 8.3. These results indicate a possible change in crystal growth kinetics during struvite crystallization. This might be related to a variation of the surface charge, or surface influence, from the increased amounts of H2PO 4 at a pH of around 7, which was also reported in the study (Ronteltap et al., 2010). Moreover, XRD results (Fig. S2) show larger presence intensity of newberyite (HMgPO43H2O) in the assay at a pH of 8.3 than in the assay of a pH 9.0. Thus, the splits of peaks at pH values of 6.7 and 8.3 might be related to the formation of newberyite because the formation of struvite and newberyite is also as function of pH (Abbona et al., 1982).
3.4. Effect of temperature on solubility product and particle size distribution of struvite under the marine/brackish condition The equilibrium activities of Mg2+, NH+4 and PO3 4 of the assays (No. 10–13) were calculated using PHREEQC. The trend of ion activity product (IAP), i.e. product of activities of Mg2+, NH+4 and PO3 4 , as function of temperature shown in Fig. S3 agreed well with the results reported by Aage et al. (1997) and Burns and Finlayson (1982). That is: the solubility product of struvite increases over temperature. In contrast, Bhuiyan et al. (2007) reported that struvite solubility product increased with temperature in the range of 10 to 35 °C and decreased in the range from 35 to 60 °C. They hypothesized that this was due to the transition of struvite, that is, from struvite (NH4MgPO46H2O) to dittmarite (NH4MgPO4H2O). However, the XRD results of crystals formed at a temperature of 46 °C in No. assay 12 and at a temperature of 56 °C in No. 13 in this present study show that the crystals were NH4MgPO46H2O and newberyite, and no dittmarite was observed, possibly explaining the difference between the results in this current study and those of Bhuiyan et al. (2007). Fig. 4 shows that crystal size increased with the increase in temperature, and also the distribution seems to be relatively more even at 46 and 56 °C than at 4 °C. Temperature substantially influences crystal growth rate (Bhuiyan et al., 2008). At higher temperature, struvite has higher solubility, and the solution became less supersaturation, which is favorable for struvite crystal growth, rather than for nucleation (Desmidt et al., 2013; Ronteltap et al., 2007). In addition, at 46 and 56 °C, two peaks in the PSD were observed, which was also observed in the study of Ronteltap et al. (2010). This may be related to the co-existence of struvite and newberyite as the XRD results shown in Fig. S2.
1155
3.5. Comparison of composition of crystals formed in synthetic brackish water and digestate filtrate XRD analyses (Fig. S4) show that the dominant crystals formed from digestate filtrate consisted of N-struvite that contains ammonia and dittmarite, whereas the crystals of synthetic brackish water consisted of struvite and newberyite (Fig. S2). That difference in composition is likely related to the difference in the initial NH4-N concentrations, that is: 179 mM in the synthetic brackish water test No. 6 and 220 mM in the filtrate. The final equilibrium NH4-N concentrations in the synthetic brackish water and the filtrate were 177 mM and 208 mM, respectively. The higher concentration of ammonia likely enhanced the formation of ammonia compounds such as struvite and dittmarite instead of the amorphous magnesium and calcium phosphates (Crutchik and Garrido, 2011; Crutchik et al., 2013). In addition, in this study Kstruvite that contains potassium instead of ammonia was not observed based on the XRD results probably due to the relative low concentration of potassium ranging from 5.12 to 19.2 mM, compared to the concentration of ammonia (Table 1). It is pronounced that the characteristics of (waste)waters played also an important role on the nature of precipitates containing ammonia and phosphate, which was also observed by Crutchik et al. (2013).
4. Conclusions In this study, it can be concluded that: 1) In the digester, struvite was formed due to the naturally present magnesium and the released ammonia and phosphate, which kept phosphate concentration low in the digester; the digester with pH of 7.7 delivered proper condition for struvite formation. 2) Estimated thermodynamic solubility product (K osp ) and estimated enthalpy change (DHhr ) of struvite formation under and the marine/brackish condition are 1013.06 o 1 25.7 kJ mol , respectively. The estimated K sp and DHhr coincide with the literature values obtained under high salinity conditions. The estimated parameters of struvite formation can be used to predict struvite precipitation in marine/ brackish aquaculture systems and anaerobic reactors treating the salty RAS sludges. 3) The average crystal size of precipitates under the marine/ brackish condition decreased with pH. 4) The average crystal size increased with temperature under the marine/brackish condition. 5) High NH+4 enhanced formation of the ammonia precipitates struvite and dittmarite.
Acknowledgement The authors acknowledge EM-MARES project and Agentschap NL for financial support to this research and China Scholarship Council for granting scholarships to Xuedong Zhang.
Appendix A. Supplementary data Fig. 4. The particle size distribution of the tests No. 6, 10, 11, 12 and 13 with various temperatures.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2016.07. 088.
1156
X. Zhang et al. / Bioresource Technology 218 (2016) 1151–1156
References Aage, H.K., Andersen, B.L., Blom, A., Jensen, I., 1997. The solubility of struvite. J. Radioanal Nucl. Chem. 223 (1–2), 213–215. Abbona, F., Lundager Madsen, H.E., Boistelle, R., 1982. Crystallization of two magnesium phosphates, struvite and newberyite: effect of pH and concentration. J. Cryst. Growth 57 (1), 6–14. Bhuiyan, M.I.H., Mavinic, D.S., Beckie, R.D., 2008. Nucleation and growth kinetics of struvite in a fluidized bed reactor. J. Cryst. Growth 310 (6), 1187–1194. Bhuiyan, M.I.H., Mavinic, D.S., Beckie, R.D., 2007. A solubility and thermodynamic study of struvite. Environ. Technol. 28 (9), 1015–1026. Burns, J.R., Finlayson, B., 1982. Solubility product of magnesium ammonium phosphate hexahydrate at various temperatures. J. Urol. 128 (2), 426–428. Cordell, D., Drangert, J.O., White, S., 2009. The story of phosphorus: global food security and food for thought. Global Environ. Change 19 (2), 292–305. Crutchik, D., Garrido, J.M., 2011. Struvite crystallization versus amorphous magnesium and calcium phosphate precipitation during the treatment of a saline industrial wastewater. Water Sci. Technol. 64 (12), 2460–2467. Crutchik, D., Sanchez, A., Garrido, J.M., 2013. Simulation and experimental validation of multiple phosphate precipitates in a saline industrial wastewater. Sep. Purif. Technol. 118, 81–88. Desmidt, E., Ghyselbrecht, K., Monballiu, A., Rabaey, K., Verstraete, W., Meesschaert, B.D., 2013. Factors influencing urease driven struvite precipitation. Sep. Purif. Technol. 110, 150–157. Hanhoun, M., Montastruc, L., Azzaro-Pantel, C., Biscans, B., Freche, M., Pibouleau, L., 2011. Temperature impact assessment on struvite solubility product: a thermodynamic modeling approach. Chem. Eng. J. 167 (1), 50–58. He, Z.W., Liu, W.Z., Wang, L., Yang, C.X., Guo, Z.C., Zhou, A.J., Liu, J.Y., Wang, A.J., 2016. Role of extracellular polymeric substances in enhancement of phosphorus release from waste activated sludge by rhamnolipid addition. Bioresour. Technol. 202, 59–66. Lahav, O., Telzhensky, M., Zewuhn, A., Gendel, Y., Gerth, J., Calmano, W., Birnhack, L., 2013. Struvite recovery from municipal-wastewater sludge centrifuge
supernatant using seawater NF concentrate as a cheap Mg(II) source. Sep. Purif. Technol. 108, 103–110. Le Corre, K.S., Valsami-Jones, E., Hobbs, P., Parsons, S.A., 2009. Phosphorus recovery from wastewater by struvite crystallization: a review. Crit. Rev. Environ. Sci. Technol. 39 (6), 433–477. Liu, B.X., Giannis, A., Zhang, J.F., Chang, V.W.C., Wang, J.Y., 2013. Characterization of induced struvite formation from source-separated urine using seawater and brine as magnesium sources. Chemosphere 93 (11), 2738–2747. Matynia, A., Koralewska, J., Wierzbowska, B., Piotrowski, K., 2006. The influence of process parameters on struvite continuous crystallization kinetics. Chem. Eng. Commun. 193 (2), 160–176. Nelson, N.O., Mikkelsen, R.L., Hesterberg, D.L., 2003. Struvite precipitation in anaerobic swine lagoon liquid: effect of pH and Mg: P ratio and determination of rate constant. Bioresour. Technol. 89 (3), 229–236. O’Neal, J.A., Boyer, T.H., 2013. Phosphate recovery using hybrid anion exchange: applications to source-separated urine and combined wastewater streams. Water Res. 47 (14), 5003–5017. Ronteltap, M., Maurer, M., Gujer, W., 2007. Struvite precipitation thermodynamics in source-separated urine. Water Res. 41 (5), 977–984. Ronteltap, M., Maurer, M., Hausherr, R., Gujer, W., 2010. Struvite precipitation from urine – influencing factors on particle size. Water Res. 44 (6), 2038–2046. Stumm, W., Morgan, J.J., 1996. Aquatic Chemistry: Chemical Equalibria and Rates in Natural Waers, third ed. Wiley Interscience, New York. Vaccari, D.A., 2009. Phosphorus: a looming crisis. Sci. Am. 300 (6), 54–59. Xu, Y.F., Hu, H., Liu, J.Y., Luo, J.H., Qian, G.R., Wang, A.J., 2015. PH dependent phosphorus release from waste activated sludge: contributions of phosphorus speciation. Chem. Eng. J. 267, 260–265. Zhang, X., Spanjers, H., van Lier, J.B., 2013. Potentials and limitations of biomethane and phosphorus recovery from sludges of brackish/marine aquaculture recirculation systems: a review. J. Environ. Manage. 131, 44–54. Zhang, X., Tao, Y., Hu, J., Liu, G., Spanjers, H., van Lier, J.B., 2016. Biomethanation and microbial community changes in a digester treating sludge from a brackish aquaculture recirculation system. Bioresour. Technol. 214, 338–347.