1 The solubility of calcium phosphate in

2 concentrated dairy effluent brines

3 K. Kezia1, J. Lee2, B. Zisu3, G. Q. Chen1, S. L. Gras1,4, S. E. Kentish1*

4 1. The ARC Dairy Innovation Hub, Department of Chemical & Biomolecular Engineering,

5 University of Melbourne, Vic 3010 Australia

6 2. Department of Chemical and Process Engineering, University of Surrey Guildford, Surrey

7 GU27XH United Kingdom

8 3. School of Applied Science, College of Science, Engineering and Health, RMIT University

9 Melbourne, Vic 3000 Australia

10 4. The Bio21 Molecular Science and Biotechnology Institute, University of Melbourne, Vic

11 3010 Australia

12

13 Keywords: calcium phosphate; dairy; brine; solid phase equilibria; lactate; citrate.

14

15 Corresponding Author

16 *Email: [email protected] Phone: +61 3 8344 6682

17
18

1
19 ABSTRACT

20 The solubility of calcium phosphate in concentrated dairy brine streams is important in

21 understanding mineral scaling on equipment such as membrane modules, evaporators and

22 heat exchangers, and in brine pond operation. In this study, the solubility of calcium

23 phosphate has been assessed in the presence of up to 300 g/L sodium chloride as well as

24 lactose, organic acids and anions at 10oC, 30oC and 50oC. As a neutral molecule, lactose has a

25 marginal, but still detectable effect upon calcium solubility. However, additions of sodium

26 chloride up to 100 g/L result in a much greater increase in calcium solubility. Beyond this

27 point, the concentrations of ions in the solution decreases significantly. These changes in

28 calcium solubility can readily be explained through changes in the activity coefficients. There

29 is little difference in calcium phosphate speciation between 10 and 30oC. However, at 50oC,

30 the ratio of calcium to phosphate in the solution is lower than at the other temperatures and

31 varies less with ionic strength. While the addition of sodium lactate has less effect upon

32 calcium solubility than sodium citrate, it still has a greater effect than sodium chloride at an

33 equivalent ionic strength. Conversely, when these organic anions are present in the solution

34 in the acid form, the effect of pH dominates and results in much higher solubility and a

35 calcium/phosphate ratio close to one, indicative of dicalcium phosphate dihydrate as the

36 dominant solid phase.

37

2
38 1. INTRODUCTION

39 With increasingly stringent environmental regulation, dairy industries in Australia are under

40 pressure to pretreat saline effluent so as to improve the quality and reduce the volume of

41 discharged effluent. This effluent stream comprises small amounts of milk sugar, large

42 concentrations of sodium chloride (NaCl) from cheese manufacturing plants, as well as

43 smaller amounts of divalent salts such as calcium phosphates. Due to the low economic value

44 of the discharged effluent, any concentration process should be performed at minimum cost.

45 Such processes are commonly conducted by membrane filtration or thermal evaporation.

46 During the concentration process, heterogenous precipitation of minerals on the surface of

47 evaporators or membrane modules often occurs, creating a major fouling issue and reducing

48 the lifetime of the equipment. If the salt concentration exceeds the limits set by local water

49 authorities, it may then be necessary to divert the saline effluent to a brine pond, where

50 further precipitation occurs.

51 The primary precipitating agent is calcium phosphate, which possesses low solubility in

52 comparison to sodium chloride or lactose. Studies of the solubility of calcium phosphate are

53 complex and ongoing, due to incongruent dissolution phenomenon and dynamic intermediate
1-4
54 equilibria . Calcium phosphate is a biomineral, which can exhibit various forms with

55 different ratios of calcium to phosphate. The solubility of calcium phosphate is also affected

56 by pH, temperature, the presence of ionic impurities and their concentration, the total ionic
5, 6
57 strength and the precipitation rate . Compared to the other parameters, pH has the most

58 profound effect and has been extensively studied 7-9.

59 The solubility of different calcium phosphate compounds across variable pH ranges are listed

60 in Table 1. In cheese whey where the pH is typically 4 – 5, dicalcium phosphate dihydrate

61 (DCPD) is the most common precipitant. At higher pH, salts such as octacalcium phosphate

3
62 (OCP) or hydroxyapatite (HAP) are more thermodynamically stable (Table 1). However in

63 dairy processing, DCPD often remains dominant. This is usually attributed to faster

64 crystallization kinetics for DCPD10 11, 12

. Inhibition of HAP and subsequent formation of

65 DCPD has also been attributed to the presence of citrates in these dairy systems 13, 14.

66 At room temperature, phosphoric acid dissociates based on the pH of the environment, as

15
67 shown below . This ionization governs the cation pairing mechanism to maintain electro-

68 neutrality within the solution.

69 ↔ 2.1 (1)

70 ↔ 7.2 (2)

71 ↔ 12.3 (3)

72 In dairy systems such as saline effluent and salty whey permeate, the solubility is also greatly

73 affected by the presence of background electrolytes. In particular, high concentrations of

74 NaCl and organic anions such as lactate and citrate are often present. For example, in a saline

75 effluent such as salty whey permeate, the concentration of citrate ranges from 0.8 mM to 3.6

76 mM. Citrate is known to sequester calcium to form the soluble calcium citrate anion (CaCit-),
16, 17
77 which increases calcium solubility . In fresh salty whey permeate, lactate concentration

78 can vary from 1.6 mM to 37 mM. This lactate concentration can further increase with time

79 due to the action of residual lactic acid bacteria which convert lactose into this acid. Indeed,

80 we have observed lactate concentrations of over 100 mM for samples stored in our laboratory

81 for six months.

82 The initial amount of NaCl can range between 30 g/L to 100 g/L (0.5 – 1.7M) and this could

83 be concentrated up to an equivalent of 250-300 g/L (> 4 - 5.1M) during effluent treatment.

84 Similarly, the calcium concentration might start at 40mM (7 g/L DCPD) but increase to 120

4
85 mM (20 g/L DCPD) during concentration18. Further, process temperatures can swing between

86 production and effluent treatment steps, ranging from 4oC to greater than 50oC. There is

87 currently little information regarding the solubility and precipitation behavior of calcium

88 phosphate under such operating conditions.

89 The current study aims to deepen our understanding of these processes and to assess the

90 possibility for selective precipitation of calcium phosphate salt. To achieve this objective, we

91 mimic the impurities and ionic strength of saline effluent in a full-scale system.

92 Understanding the precipitation of calcium phosphate in the presence of extreme

93 concentrations of NaCl and organic anions and acids will aid in mitigating mineral scaling

94 and thus prolong the lifetime of equipment through the adjustment of operating conditions.

95 These findings have applications beyond the treatment of dairy saline effluent, encompassing

96 water treatment processes within other chemical industries.

97

5
 98 2. MATERIAL AND METHODS

99 Materials

100 All chemicals used were analytical grade and were used as received without further

101 purification. For experiments and cleaning procedures, purified water (Elix Millipore,

102 resistivity > 15 MΩ cm-1) was used. For analytical procedures and preparation of standard

103 solutions, water of higher purity (Milli-Q Millipore, resistivity > 18 MΩ cm-1) was used.

104 Dicalcium phosphate dihydrate, DCPD (CaHPO4.2H2O, >98%) was obtained from Astral

105 Scientific. Calcium chloride dihydrate (CaCl2.2H2O, >99%), di-Sodium hydrogen phosphate

106 (Na2HPO4, >99%) and lactose monohydrate (C12H22O11.H2O) were purchased from Chem

107 Supply. To adjust the pH and organic anion concentration, tri-sodium citrate (Na3Cit, >99%),

108 sodium lactate (NaC3H5O3, >70%), citric acid (C6H8O7, >99%) and lactic acid (C3H6O3,

109 85%) were purchased from Chem Supply and nitric acid (HNO3, 69.5%) was purchased from

110 Scharlau.

111 Methods

112 The experiments were performed using an end-point equilibrium technique. This technique

113 operates by incubating excess amounts of solid in a pre-conditioned background solution at

114 constant temperature. It has been widely utilized for solubility measurement 19. In the present

115 case, a consistent excess amount (2 g unless otherwise stated) of solid DCPD (Ca/P =1) was

116 added to 200 ml of a background solution in a glass container, as shown in Figure 1. The

117 composition of the background solution was adjusted to be between 0 – 300 g/L (5.1 M)

118 NaCl and 0 to 100 g/L (0.3M) of lactose. The role of organic acid and anions was determined

119 using a background solution of 1 to 100 mM of either lactic or citric acid, sodium lactate or

120 sodium citrate. The vessel was double-sealed using Parafilm M,R Laboratory film (SPI

6
121 supplies) and capped to avoid evaporation. It was then incubated in a circulating water bath

122 (Julabo) controlled at 10oC, 30oC or 50oC for 168 hours (one week). A constant stirring rate

123 was set at approximately 250 rpm. At the end of the incubation period, the supernatant was

124 filtered using a 0.2 µm (polyethersulfone, PES) syringe filter (Millipore) and the composition

125 of the supernatant liquor was analysed. The pH was measured at the beginning and at the end

126 of the experiment, using a WP81 double junction glass pH electrode and temperature probe,

127 connected to a digital benchtop meter. Each experiment was carried out at least in duplicate.

128 At the conclusion of experiments, the concentration of Ca, Na and phosphorus (P) was

129 measured using Inductively Coupled Plasma – Optical Emission Spectrometry (ICP-OES)

130 (Varian 720-OES). The concentrations of the orthophosphate anions (H2PO4-, HPO42-, PO43-)

131 and chloride (Cl-) were determined using Ion Chromatography (Dionex 1000CS) with an AS

132 14 anion exchange column. The concentration of lactate and citrate was quantified using

133 Reverse Phase High Performance Liquid Chromatography (RP-HPLC) (Shimadzu LC-20AT)

134 employing an ion-exchange column (HPX-87H). Triplicate measurements were performed

135 for each independent sample and the error margins are based on a single standard deviation of

136 these results (n=6). Data was analysed using one way analysis of variance (ANOVA), with a

137 significance level of p = 0.01.

138 The behavior of calcium and phosphate ions in the solution was also simulated through

139 predictions of the activity coefficient of these ions at ionic strengths of 0 to 5M NaCl.

140 ASPEN Plus V8.6 (Aspen One) was employed using the Pitzer thermodynamic package.

141

7
142 3. RESULTS AND DISCUSSION

143 Effect of neutral solute (lactose)

144 It was expected that lactose would not alter the solubility of calcium phosphate, as a neutral

145 solute should not dictate ion speciation phenomena. However, there was a significant increase

146 (p < 0.01) in the concentration of soluble calcium from 0.68 ± 0.06 mM to 1.33 ± 0.10 mM,

147 as the lactose concentration increased from 0 to 100 g/L (Figure 2). It can also be seen from

148 Figure 2b that the calcium to phosphate ratio (Ca/P) in the supernatant in the absence of

149 lactose or NaCl is significantly less than unity (p<0.01), indicating that the solid phase has a

150 Ca/P ratio greater than unity, indicative of significant quantities of HAP or OCP (Table 1).

151 As the amount of the remaining excess solid in the solution is significantly greater than the

152 amount dissolved, an accurate mass balance of the solid could not be conducted to confirm

153 the exact solid phase composition.

154 Increasing the lactose concentration significantly increases (p < 0.01) the Ca/P ratio, from

155 0.54 ± 0.01 to 0.70 ± 0.01 mM. The effect of sugars on different calcium salts has also been
20 21
156 observed by Doherty et al. in a calcium oxalate-sucrose system. Besic claims that

157 sucrose in food products affects the solubility of calcium and phosphate in teeth. Other

158 workers have shown that calcium salts can significantly influence the solubility and growth
22, 23
159 rate of lactose crystals due to the formation of a lactose complex . The changing

160 solubility may also reflect a decrease in the water activity as the lactose concentration

161 increased, consistent with the observation of a very small conductivity increase. However,

162 while lactose clearly impacts calcium solubility, the effect is very small, when compared with

163 NaCl, as also seen in Figure 2(a). The remainder of this work therefore focuses on the impact

164 of such ionic species.

165

8
166 Effect of ionic strength

167 It can be seen from Figure 3 that the concentration of the calcium ion in the supernatant

168 increases significantly for all temperatures as the NaCl concentration increases to 100g/L.

169 Beyond this point, a significant (p <0.01) decrease in both the soluble calcium and phosphate

170 concentration is observed. This decreased solubility of calcium in solution at the highest

171 NaCl concentrations is of significance in understanding fouling in equipment such as brine

172 evaporators, which may operate under such conditions. The decreased solubility could cause

173 rapid calcium scaling to occur.

174 These trends in solubility can be understood by reference to the activity coefficient of the

175 ions in solution. With increasing ionic strength, the activity coefficient of both calcium and

176 phosphate ions decrease (Figure 4) resulting in increased solubility. This loss in activity
24-26
177 results from ion-pairing (association), which has been previously investigated , up to the

178 ionic strength of seawater (approximately 35 g/L or 0.7M) at room temperature. Millero et

179 al26 uses the Pitzer equation to predict that the activity coefficient of Ca2+ should decrease

180 from 0.851 at an ionic strength of 0.002 M down to 0.225 at an ionic strength of 0.7 M at 25
o
181 C. Their study also predicts that the activity coefficients for H2PO4- decrease from 0.947 to

182 0.363 and for HPO42- from 0.758 to 0.044. Comparable values were generated in the present

183 case, as shown in Figure 4. However, beyond a salt concentration of 100 g/L the activity

184 coefficients for calcium increase again.

185 Prediction of the supernatant concentration using these activity coefficients is a complex

186 problem given the solid phase equilibria between OCP, HAP and DCPD, in addition to the

187 phosphate speciation given by Equations 1 to 3. Figure 5 shows a simulation for the simplest

188 case where DPCD is assumed to be the only solid phase, using the solubility constant

189 provided in Table 1 at 25oC. It is clear that even this simple simulation provides a reasonable

9
190 fit to the supernatant calcium concentration. The phosphate concentrations are higher, again

191 suggesting that the solid phase contains OCP or HAP in addition to DCPD.

192 With increasing ionic strength a small decrease in pH was observed. The pH fell from 7.67

193 ± 0.25 in the absence of NaCl to 7.00 ± 0.22 at 300 g/L NaCl. At comparable pH values and

194 at zero NaCl concentration, the Ca/P ratio observed (Table 2) is similar to the value reported

195 by Sutter et al. from their leaching experiments 27. However, with increasing ionic strength,

196 the Ca/P ratio increases significantly at both 10 oC and 30 oC, with values in the range of

197 0.72-0.77 at 10 to 300 g/L NaCl. This suggests a shift back from OCP or HAP towards

198 DCPD, consistent with the common observations in dairy processing. Conversely, there is no

199 significant shift in the Ca/P ratio at 50oC, implying that at this temperature, the crystal

200 morphology is unaffected by ionic strength. Figure 4(b) also shows that the ionic strength

201 affects the activity of H2PO4-, HPO42- and PO43- to differing extents, which alters the

202 equilibria of these species. These changes may also explain the observed pH variation.

203
204 Some workers have argued that the amount of excess solid existing in the system alters the

205 solubility of calcium phosphate, with the lowest solubility occurring for the smallest amount

206 of solid 1, 28, 29. This is consistent with the present results shown in Figure 6. In the absence of

207 NaCl, no significant difference could be observed (p <0.01), as the amount of [Ca2+]

208 dissolved in the solution is very low. However, at higher ionic strength, it can be seen that the

209 solubility is significantly affected (p<0.01) by the amount of remaining excess solid in the

210 solution. These differences have been reported to relate to the different calcium phosphate

211 polymorphs that can form and the resulting equilibria between multiple solid phases, in

212 addition to the liquid-solid phase equilibria. These authors argue that this is the reason that

213 the solubility data reported by researchers using different techniques often shows wide

214 variability.

10
215

216 Effect of Temperature

217 The solubility of both calcium and phosphate is identical within experimental error at 10 oC

218 to 30 oC for NaCl concentrations of up to 50 g/L (Figure 3). However, beyond this

219 concentration, the solubility at 30oC is significantly lower than that at 10oC (p<0.01).

220 Changes in solubility with temperature are more apparent at 50oC, with the solubility

221 increasing significantly (p<0.01) at all NaCl concentrations. Sutter et al. 27 observed a slight

222 decrease in the concentration of soluble calcium in the solution from temperatures of 5 oC to

223 37 oC using a leaching technique (at comparable pH ~ 6.5-6.8) and from 25oC to 37oC in

224 batch experiments at near neutral and acidic conditions. On the other hand, Green and Perry
30
225 report an increasing solubility of monobasic calcium phosphate with respect to

226 temperature. These literature results are all broadly consistent with the work presented here.

227 However, the temperature effects in this case should not be oversimplified as the Ca/P ratio in

228 the solution is also significantly altered, especially at 50 oC where this ratio decreases to 0.45

229 ± 0.05 (Table 2).

230 The results appear initially inconsistent with many previous studies, which claim that calcium
31-33
231 phosphate solubility declines with temperature . Indeed, in our own work, we have also

232 previously observed that increasing the temperature could shorten the time it takes for the

233 first appearance of crystals in saline dairy effluent34. However, these references generally

234 describe kinetic effects, where increasing temperature causes calcium salts to precipitate in a

235 shorter time. The present work and that of the literature cited above considers the final

236 equilibrium after at least 168 hours, where such kinetic effects become irrelevant.

237
238

11
239 Effect of Organic anions

240 Figure 6 shows the concentration of calcium in the solution in the presence of either the

241 organic lactic and citric acids, or the strong acid HNO3. The increasing solubility of calcium

242 phosphate with respect to pH is well known (pH < 6), although this effect varies depending

243 upon the pKa of the specific acid. The role of pH in governing the solubility of calcium

244 phosphate salts by addition of orthophosporic acid has been reported . However,
35, 36

245 orthophosphoric acid exhibits both weak acid and polybasic properties, and will alter the

246 amount of available phosphate in the solution through the common ion effect. In this study,

247 HNO3 is used to isolate the role of pH, as the formation of Ca(NO3)2 is unlikely to occur.

248 In 1 mM HNO3, lactic acid or citric acid, the initial pH of the solution (with 20 g/L DCPD)

249 is 4.61 ± 0.14, 5.67 ± 0.38 and 5.00 ± 0.33, respectively. Within this pH range the

250 concentration of calcium recorded in solution is comparable to the value reported by Sutter et
27
251 al. obtained using a leaching method where the pH was adjusted using H3PO4. The values

252 recorded at 1 mM are also comparable with that obtained by other workers using OCP as the
37
253 starting solid . The calcium solubility is lowest for the lactic acid (p<0.01), reflecting its

254 weak acidity and hence the relatively high pH of the solution. The solubility of the nitric and

255 citric acids are comparable within experimental error, even though the pH of the citric

256 solution is higher than that of the nitric. This reflects the calcium sequestering capacity of the

257 citrate anion.

258 Discrepancies with the literature are observed at higher acidity (10 mM, see Figure 7). The

259 solubility of calcium in the solution is two to four times lower compared to the concentration

260 reported by Sutter et al. 27. At the end of experiments, these workers obtain a value of 47mM

261 at similar pH (pH 2.90 and 3.53 for HNO3 and lactic acid respectively in this work, versus

12
262 3.38 in the Sutter et al. study). This might be due to the utilisation of different acids here

263 instead of orthophosphoric acid.

264 Figure 8a shows that the solubility of calcium ions increases significantly (p<0.01) as the

265 sodium lactate concentration increases from 1 to 100 mM at 10 and 30 oC. The use of sodium

266 lactate in these experiments ensures that the pH of the solution changes little, within the range

267 of pH 7.0 to 8.7. The increased solubility is partly associated with the increase in the

268 background ionic strength. Nevertheless, the effect of sodium lactate on the solubility of

269 calcium and phosphate is more pronounced than NaCl at an equivalent concentration.

270 Specifically, the effect of 100 mM sodium lactate on calcium solubility is equivalent (p>0.05)

271 to 170 mM (10g/L) of NaCl at 10 oC and 30 oC (see Table 3).

272 At 50oC, the pH falls significantly upon addition of sodium lactate, suggesting that

273 different calcium phosphate speciation is occurring. The calcium solubility increases further,

274 but it is difficult to confirm whether this is due to the lactate or the shift in pH.

275 In dairy effluent, the amount of citrate is relatively low compared to lactate. However,

276 much greater increases in the concentration of calcium in the solution are observed when the

277 citrate anion is added to solution (Figure 8a). The contribution of pH is again minimized here

278 through utilization of tri-sodium citrate (pH 7.3-8.2). It is clear that citrate shows a strong

279 calcium sequestering capacity, as has been widely reported elsewhere 4, 35, 38-43. The effect of

280 temperature on the saturation concentration of calcium in the presence of the citrate anion is

281 less clear than with lactate or chloride, due to the dominant effect of citrate sequestration

282 (Figure 8a).

283 A steady increase in the calcium concentration in the solution is observed for both organic

284 acids as shown in Figure 8b. The sequestering capacity of citrate dominates over the

285 contribution of pH under all conditions, with the total calcium in the citric acid solution three

13
286 to four times higher than in lactic acid. Further, analysis indicated that after 168 hours

287 equilibration, from an initial concentration of 10 mM of organic acid, 5.4 ± 0.8 mM of citrate

288 was consumed whereas only 1.5 ± 0.3 mM lactate was consumed for the same amount of

289 DCPD. At 100mM (Figure 6), the concentration of calcium in the solution is largely the

290 same for lactic and nitric acid, which confirms that the increase in solubility for lactic acid is

291 solely due to phosphate dissociation at these extreme pH levels (DCPD in 100mM HNO3 pH

292 1.75 and in 100 mM lactic acid pH 2.68).

293 The addition of sodium lactate has minimal impact on the Ca/P ratio at 30oC (see Figure 9).

294 Conversely, the addition of trisodium citrate increases the amount of the available calcium in

295 the solution due either to formation of CaCit- or to the preferential formation of DCPD 13, 14 .

296 The addition of lactic acid causes the Ca/P ratio to increase to unity. This is consistent with

297 the literature; DCPD is the most stable salt at pH less than 5.5 (Table 1). A similar increase

298 was expected for citric acid, however this was not clearly observed, with the Ca/P in this case

299 showing significant variability (data not shown).

300 As with the systems containing NaCl (Table 2) in systems at neutral pH, high temperature

301 (50 oC) results in a lower Ca/P ratio in solution (p<0.01) implying preferential formation of

302 HAP or OCP (Table 4). In the presence of lactic acid, the Ca/P ratio remains close to unity

303 regardless of the temperature. This shows the dominating effect of pH in comparison to other

304 parameters. Conversely, with citric acid, this ratio falls with temperature (p < 0.01).

305

306 Effect of high ionic strength in the presence of lactic acid

307 In real saline waste effluent, the background electrolyte comprises a mixture of ions. Figure

308 10 shows the saturation concentration of calcium in a mixture of NaCl with 10 mM of lactic

14
309 acid at 30 oC. The pH range throughout the whole experiment was 3.7 ± 0.6. It can be seen

310 that the trends in solubility observed with NaCl alone (Figure 3) are no longer observed.

311 Instead, the concentration of calcium in the solution remains relatively constant at the same

312 value as for 10mM lactic acid alone (7 mM). This again reflects the stronger influence of pH,

313 relative to those of ionic strength alone.

314 Despite the dominance of pH in determining the calcium solubility, the presence of NaCl

315 still significantly affects the Ca/P ratio. In the absence of NaCl, the Ca/P ratio observed is

316 1.01 ± 0.03 which is consistent with the Ca/P ratio across a broad range of acidic conditions.

317 However, in the presence of NaCl the Ca/P ratio falls significantly (p<0.01) to 0.8 ± 0.03.

318 4. DISCUSSION

319 In this work, the solubility of calcium phosphate in the presence of high concentrations of

320 lactose, sodium chloride, lactate and citrate and under conditions of variable acidity and

321 temperature has been investigated. Lactose has a marginal but significant effect upon the

322 saturation concentration of calcium in the solution, possibly due to the formation of a

323 complex of lactose with calcium salts. Increasing ionic strength through the addition of NaCl

324 up to 100 g/L within the temperature range of 10oC to 50oC, reduces the activity of Ca2+

325 resulting in an increase of calcium in the solution. Above this NaCl concentration, the

326 calcium solubility declines, particularly at 50oC. This latter finding is of importance in

327 understanding calcium scaling in high salt solutions. The addition of salts such as sodium

328 chloride or trisodium lactate has differing effects on the solubility of H2PO4- and HPO42- at

329 different temperatures, so that at 10 and 30oC, the Ca/P ratio increases from around 0.5 to

330 0.7-0.8, while at 50oC this ratio remains low.

331 The lactate anion shows only a very small calcium sequestering capacity compared to the

332 citrate anion, but is still a more effective sequestering agent than chloride. Specifically, at

15
333 10oC and 30oC, the same calcium solubility is achieved by 100 mM sodium lactate as 170

334 mM sodium chloride. However, when lactate is added as an organic acid, the role of pH

335 dominates, providing large increases in the calcium concentration and resulting in the Ca/P

336 ratio reaching unity.

337 In a real effluent solution comprising an abundant amount of both NaCl and lactic acid, the

338 effect of pH dominates the saturation concentration of calcium, but crystal morphology is still

339 affected by the NaCl concentration, as indicated by changes in the Ca/P ratio.

340 Lastly, experiments showed some evidence in support of previous work by other authors

341 that have suggested that calcium solubility can be affected by the quantity of excess solid in

342 the solution.

343
344 5. ACKNOWLEDGMENTS

345 This research was supported under Australian Research Council’s Industrial

346 Transformation Research Program (ITRP) funding scheme (Project number IH120100005).

347 The ARC Dairy Innovation Hub is a collaboration between The University of Melbourne,

348 The University of Queensland and Dairy Innovation Australia Ltd. Judy Lee acknowledges

349 the support from an Australian Research Council Discovery Early Career Researcher Award

350 (DE120101567).

351
352 REFERENCES

353 1. Pan, H. B.; Darvell, B. W., Calcium Phosphate Solubility: The Need for Re-
354 Evaluation. Cryst. Growth Des. 2009, 9, 639-645.
355 2. Miyazaki, T.; Sivaprakasam, K.; Tantry, J.; Suryanarayanan, R., Physical
356 characterization of dibasic calcium phosphate dihydrate and anhydrate. J. Pharm. Sci. 2009,
357 98, 905-916.
358 3. Pan, H. B.; Darvell, B. W., Solubility of Dicalcium Phosphate Dihydrate by Solid
359 Titration. Caries Res. 2009, 43, 254-260.

16
360 4. Spanos, N.; Patis, A.; Kanellopoulou, D.; Andritsos, N.; Koutsoukos, P., Precipitation
361 of calcium phosphate from simulated milk ultrafiltrate solutions. Cryst. Growth Des. 2007, 7,
362 25-29.
363 5. Dorozhkin, S., Calcium Orthophosphates: Occurrence, Properties and Major
364 Applications. Bioceram. Dev. Appl. 2014, 4, 2.
365 6. Wang, L.; Nancollas, G. H., Calcium orthophosphates: crystallization and dissolution.
366 Chem. Rev. 2008, 108, 4628-4669.
367 7. Johnsson, M. S.-A.; Nancollas, G. H., The Role of Brushite and Octacalcium
368 Phosphate in Apatite Formation. Crit. Rev. Oral Biol. & Med. 1992, 3, 61-82.
369 8. Lundager Madsen, H. E.; Thorvardarson, G., Precipitation of calcium phosphate from
370 moderately acid solution. J. Cryst. Growth 1984, 66, 369-376.
371 9. Madsen, H. E. L.; Christensson, F., Precipitation of calcium phosphate at 40 C from
372 neutral solution. J. Cryst. Growth 1991, 114, 613-618.
373 10. Ferreira, A.; Oliveira, C.; Rocha, F., The different phases in the precipitation of
374 dicalcium phosphate dihydrate. J. Cryst. Growth 2003, 252, 599-611.
375 11. House, W. A., The physico-chemical conditions for the precipitation of phosphate
376 with calcium. Environ. Technol. 1999, 20, 727-733.
377 12. Van Kemenade, M.; De Bruyn, P., A kinetic study of precipitation from
378 supersaturated calcium phosphate solutions. J. Colloid Interface Sci. 1987, 118, 564-585.
379 13. Rosmaninho, R.; Melo, L. F., The effect of citrate on calcium phosphate deposition
380 from simulated milk ultrafiltrate (SMUF) solution. J. Food Eng. 2006, 73, 379-387.
381 14. Van Der Houwen, J. A. M.; Valsami-Jones, E., The Application of Calcium
382 Phosphate Precipitation Chemistry to Phosphorus Recovery: The Influence of Organic
383 Ligands. Environ. Technol. 2001, 22, 1325-1335.
384 15. CRC Handbook of Chemistry and Physics. 58th ed.; CRC Press: Cleveland Ohio,
385 1978.
386 16. Mekmene, O.; Gaucheron, F., Determination of calcium-binding constants of caseins,
387 phosphoserine, citrate and pyrophosphate: A modelling approach using free calcium
388 measurement. Food Chem. 2011, 127, 676-682.
389 17. Sharma, V. K.; Johnsson, M.; Sallis, J. D.; Nancollas, G. H., Influence of citrate and
390 phosphocitrate on the crystallization of octacalcium phosphate. Langmuir 1992, 8, 676-679.
391 18. Kezia, K.; Lee, J.; Weeks, M.; Kentish, S., Direct contact membrane distillation for
392 the concentration of saline dairy effluent. Water Res. 2015.
393 19. Goss, S. L.; Lemons, K. A.; Kerstetter, J. E.; Bogner, R. H., Determination of calcium
394 salt solubility with changes in pH and PCO2, simulating varying gastrointestinal
395 environments. J. Pharm. Pharmacol. 2007, 59, 1485-1492.
396 20. Doherty, W.; Wright, P. A solubility model for calcium oxalate formation in a sugar
397 mill, in Proceedings-Australian Society Of Sugar Cane Technologists, 2004; PK Editorial
398 Services; 1999: 2004; pp 50.
399 21. Besic, F. C., Food materials for prevention of carious degradation of teeth.United
400 States Patent No. 4515770 A 1985.
401 22. Jelen, P.; Coulter, S. T., Effects Of Certain Salts And Other Whey Substances On The
402 Growth Of Lactose Crystals. J. Food Sci. 1973, 38, 1186-1189.
403 23. Jensen, O.; Hanford, Z.; Supplee, G., Equilibrium solutions of certain lactose-salt
404 mixtures. J. Dairy Sci. 1940, 23, 745-753.
405 24. Atlas, E.; Culberson, C.; Pytkowicz, R., Phosphate association with Na+, Ca2+ and
406 Mg2+ in seawater. Marine Chem. 1976, 4, 243-254.
407 25. Kester, D. R.; Pytkowicz, R. M., Theoretical model for the formation of ion-pairs in
408 seawater. Marine Chem. 1975, 3, 365-374.

17
409 26. Millero, F. J.; Schreiber, D., Use of the ion pairing model to estimate activity
410 coefficients of the ionic components of natural waters. Am. J. Sci. 1982, 282, 1508-1540.
411 27. Sutter, J.; McDowell, H.; Brown, W. E., Solubility study of calcium hydrogen
412 phosphate. Ion-pair formation. Inorg. Chem. 1971, 10, 1638-1643.
413 28. Greenwald, I., The solubility of calcium phosphate I. The effect of pH and of amount
414 of solid phase. J. Biol. Chem. 1942, 143, 703-710.
415 29. Pan, H. B.; Darvell, B. W., Solubility of hydroxyapatite by solid titration at pH 3–4.
416 Arch. Oral Biol. 2007, 52, 618-624.
417 30. Green, D.; Perry, R., Perry's Chemical Engineers' Handbook, Eighth Edition.
418 McGraw-Hill Education: 2007.
419 31. Koutsoukos, P. G. Current knowledge of calcium phosphate chemistry and in
420 particular solid surface–water interface interactions, In Proceedings of the Second
421 International Conference on Phosphorus Recovery for Recycling from Sewage and Animal
422 Wastes, Institute of Chemical Engineering and High Temperature Chemical Processes, Univ.
423 of Patras, 2000; pp 12-14.
424 32. Rice, G.; Barber, A.; O'Connor, A.; Stevens, G. W.; Kentish, S. E., Fouling of NF
425 membrane by dairy ultrafiltration permeate. J. Membr. Sci. 2009, 330, 117-126.
426 33. Schäfer, A.; Fane, A. G.; Waite, T., Nanofiltration of natural organic matter: removal,
427 fouling and the influence of multivalent ions. Desalin. 1998, 118, 109-122.
428 34. Kezia, K.; Lee, J.; Zisu, B.; Weeks, M.; Chen, G.; Gras, S.; Kentish, S.,
429 Crystallisation of minerals from concentrated saline dairy effluent. Water Res. 2016, 101,
430 300-308.
431 35. Amjad, Z. (Ed.), Calcium Phosphates in Biological and Industrial Systems. Springer
432 US: 2013.
433 36. Chow, L. C.; Eanes, E. D., Octacalcium Phosphate. Karger: 2001; Vol. 18.
434 37. Tung, M.; Eidelman, N.; Sieck, B.; Brown, W., Octacalcium phosphate solubility
435 product from 4 to 37 C. J. Res. Nat. Bur. Stand. 1988, 93, 613-624.
436 38. Mekmene, O.; Mekmene, F.; Gaucheron, Determination of calcium-binding constants
437 of caseins, phosphoserine, citrate and pyrophosphate: A modelling approach using free
438 calcium measurement. Food Chem. 2011, 127, 676-682.
439 39. Mekmene, O.; Mekmene, T.; Rouillon, S.; Quillard, P.; Pilet, J.-M.; Bouler, S.;
440 Pezennec, F.; Gaucheron, Effects of citrate and NaCl on size, morphology, crystallinity and
441 microstructure of calcium phosphates obtained from aqueous solutions at acidic or near-
442 neutral pH. J. Dairy Res. 2012, 79, 238-248.
443 40. Sharma, V.; Johnsson, M.; Sallis, J.; Nancollas, G., Influence of citrate and
444 phosphocitrate on the crystallization of octacalcium phosphate. Langmuir 1992, 8, 676-679.
445 41. Duputel, M.; Devau, N.; Brossard, M.; Jaillard, B.; Jones, D. L.; Hinsinger, P.;
446 Gérard, F., Citrate adsorption can decrease soluble phosphate concentration in soils: Results
447 of theoretical modeling. Appl. Geochem. 2013, 35, 120-131.
448 42. Holt, C., Calcium phosphate nanoclusters and their applications. United States Patent
449 No. US 7517850 B2, 2009.
450 43. Holt, C.; Jenness, R., Interrelationships of constituents and partition of salts in milk
451 samples from eight species. Comp. Biochem. Physiol. A: Mol. Integr. Physiol. 1984, 77, 275-
452 282.
453

18
454 TABLE CAPTIONS

455 Table 1. The various forms of calcium phosphate that can form in aqueous solution and their
456 solubility at 25oC and neutral pH5.

457 Table 2. The Ca/P ratio in the supernatant after 168 hours equilibration of 10 g/L DCPD, with

458 a background concentration of NaCl. Superscripts indicate samples that are not significantly

459 different (p > 0.01).

460 Table 3. The Calcium and phosphorus concentrations in the supernatant after 168 hours

461 equilibration of 10 g/L DCPD, with a background concentration of NaCl or sodium lactate.

462 Superscripts indicate samples that are not significantly different (p > 0.01).

463 Table 4. The Ca/P ratio in the supernatant after 168 hours equilibration of 10g/L DCPD, with

464 a background concentration of organic anions. Superscripts indicate samples that are not

465 significantly different (p > 0.01).

466

467

19
468 FIGURE CAPTIONS

469 Figure 1. Schematic diagram of the end-point equilibrium technique.

470 Figure 2. (a) The concentration of calcium and (b) the calcium to phosphate ratio in the

471 supernatant after 168 hours of equilibration at 30oC of 10 g/L DCPD, with a background

472 concentration of NaCl or lactose.

473 Figure 3. The concentration of calcium in the supernatant after 168 hours of equilibration

474 of 10 g/L DCPD, with a background concentration of NaCl.

475 Figure 4. (a) Activity coefficient of Ca2+ (b) Activity coefficient of orthophosphates

476 simulated using the Pitzer model within ASPEN Plus V8.6 at 25oC. Concentrations of

477 calcium, phosphate and hydronium simulated were 10mM.

478 Figure 5. Comparison of the experimental data for calcium and phosphorus concentrations

479 in the supernatant at 30oC with a simulation that assumes that DCPD is the only salt (Ca/P =

480 1) and uses the activity coefficients determined from the Pitzer model and the solubility

481 constant from Table 1 at 25oC.

482 Figure 6. Concentration of Ca2+ in the supernatant after 168 hours equilibration of 10g/L

483 DCPD and 1.25g/L DCPD at 30oC.

484 Figure 7. Concentration of calcium in the supernatant after 168 hours of equilibration of

485 20g/L DCPD at 30oC, with a background acid concentration (a) absolute concentrations (b)

486 concentrations relative to the nitric acid case.

487 Figure 8. (a) The concentration of Ca2+ in the supernatant with a background concentration

488 of an organic anion i.e. sodium lactate and trisodium citrate (pH 7.0-8.7) (b) The

489 concentration of Ca2+ in the supernatant with a background concentration of organic acid i.e.

20
490 lactic acid and citric acid (pH 2.68 - 6.00) after 168 hours of equilibration of 10g/L DCPD.

491 The NaCl concentration is zero. Note the different x-axis scales on the two graphs.

492 Figure 9. The calcium to phosphate ratio in the supernatant after 168 hours of equilibration

493 at 30oC of 10 g/L DCPD, with a background concentration of lactic acid, trisodium citrate or

494 sodium lactate.

495 Figure 10. Concentration of calcium and phosphorous in the supernatant after 168 hours of

496 equilibration at 30oC of 10g/L in 10 mM lactic acid, with a background concentration of

497 NaCl.

498

21
499 Table 1

Ca/P Compound Formula Solubility pH*

mol 25oC 25oC
ratio -log (Ksp)
0.5 Monocalcium phosphate Ca(H2PO4)2.H2O 1.14 0.0-2.0
monohydrate (MCPM)
1.0 Dicalcium phosphate dihydrate CaHPO4.2H2O 6.59 2.0-6.0
(DCPD, brushite)
1.3 Octacalcium phosphate (OCP) Ca8(HPO4)2(PO4)4.5H2O 96.6 5.5-7.0
1.2-2.2 Amorphous calcium phosphates CaxHy(PO4)z.nH2O *** 5.0-12.0
(ACP) n=3-4.5
1.5-1.7 Calcium deficient hydroxyapatite Ca10-x(HPO4)x(PO4)6- ≈ 85 6.5-9.5
(CDHA) x(OH)2-x
(0<x<1)
1.7 Hydroxyapatite (HAP) Ca10(PO4)6(OH)2 116.8 9.5-12.0
500 *pH where the solid can exist in aqueous solution at room temperature.

501 ***cannot be precisely measured, in acidic buffer ACP<CDHA<HAP.

502

22
503 Table 2

0 g/L NaCl 10 g/L NaCl 100 g/L NaCl 300 g/L NaCl
10oC 0.52 ± 0.06a 0.74 ± 0.04b 0.74 ± 0.03b 0.77 ± 0.01b

30oC 0.54 ± 0.03a 0.72 ± 0.01b 0.77 ± 0.02b 0.72 ± 0.04b

50oC 0.45 ± 0.03c 0.46 ± 0.01c 0.44 ± 0.02c 0.46 ± 0.04c

504
505

23
506 Table 3

Calcium (mM) Phosphorus (mM)

170 mM NaCl 100 mM NaLactate 170 mM NaCl 100 mM NaLactate

10oC 1.97 ± 0.13 1.99  0.22 2.65 ± 0.15 2.46  0.05

30oC 1.96 ± 0.11 1.87 ± 0.10 2.67 ± 0.06 2.71 ± 0.03

507
508

24
509 Table 4

10oC 30oC 50oC 10oC 30oC 50oC

Sodium Lactate Trisodium Citrate
0.66 ± 0.04 a 0.61 ± 0.04a 0.45 ± 0.01b 0.84 ± 0.06 c
0.81 ± 0.03c 0.27 ± 0.04d

Lactic Acid Citric Acid

0.96 ± 0.08e 1.04 ± 0.04f 0.92± 0.03 e
0.88 ± 0.07 g
0.78 ± 0.04c 0.64 ± 0.03a

510
511

512

513

25
Minerva Access is the Institutional Repository of The University of Melbourne

Author/s:
Kezia, K; Lee, J; Zisu, B; Chen, GQ; Gras, SL; Kentish, SE

Title:
Solubility of Calcium Phosphate in Concentrated Dairy Effluent Brines

Date:
2017-05-24

Citation:
Kezia, K., Lee, J., Zisu, B., Chen, G. Q., Gras, S. L. & Kentish, S. E. (2017). Solubility of
Calcium Phosphate in Concentrated Dairy Effluent Brines. JOURNAL OF AGRICULTURAL
AND FOOD CHEMISTRY, 65 (20), pp.4027-4034. https://doi.org/10.1021/acs.jafc.6b05792.

Persistent Link:
http://hdl.handle.net/11343/156394

File Description:
Accepted version