Chemical control of struvite scale by a green inhibitor polyaspartic acid

Abstract

Many efforts have been made to develop effective chemical inhibitors for struvite scale, which causes a range of operational problems in the wastewater treatment industry. Herein, the inhibitory capacity of polyaspartic acid (PASP) on the spontaneous precipitation of struvite at pH 9 was investigated. Struvite precipitates were characterized using X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM), and energy dispersive X-ray spectroscopy (EDX). Precipitation experiments dosed with PASP revealed that PASP is effective in the growth inhibition of struvite and its inhibitory capacity is proportional to its concentration and that PASP also plays a role in the morphological modification of struvite crystals. The effect of several key parameters, including pH, mixing energy, reaction time, and calcium ions, on PASP inhibition performance was examined for potential practical applications. The results showed that the inhibitory capacity of PASP is sustainable and efficient. Dissolution experiments dosed with PASP were also conducted, and the results showed that PASP can accelerate the dissolution of preformed struvite, and this capacity increases with an increase in its concentration. Therefore, PASP can potentially act as a feasible and environmentally-friendly inhibitor and cleaning agent for struvite scale.

---

1. Introduction

The formation of struvite deposits in wastewater treatment plants (WWTPs) has been widely reported since 1939 when it was first identified in digested sludge supernatant lines.¹ Struvite, which is known as magnesium ammonium phosphate hexahydrate (MgNH₄PO₄·6H₂O), crystallizes in the orthorhombic system and adopts a number of natural morphologies, including equant, wedge-shaped, short prismatic, and tabular forms.² When the concentrations of magnesium, ammonium and phosphate ions exceed the struvite solubility product, its precipitation occurs. Wastewater usually contains large amounts of phosphorus and nitrogen, and about 80% of N and 50% of P originate from urine.³ Anaerobic digestion further solubilizes organic-P and N to PO₄³⁻ and NH₄⁺, respectively, which favor the formation of struvite.^4,5 Struvite crystals tend to form a hard scale on the process equipment surfaces of WWTPs such as sludge liquors pipes, pumps, centrifuges and aerators, which leads to clogging and breakdown of these equipment.^1,6,7

To eliminate this nuisance, many efforts have been made to deal with the formation of struvite, which are based on four principal approaches. In the first approach, existing struvite scale is removed by acid washing or chipped away manually with a hammer and chisel.⁸ Obviously, this is a time and manpower consuming maintenance project. The second approach aims at reducing the potential of struvite precipitation by lowering the supersaturation level. Initially, the digested sludge stream holding high supersaturation is diluted with a secondary effluent to reduce supersaturation, but the mitigation of struvite formation was proven to be limited in practice.^9,10 Recently, the chemical dosing of iron(III) salts has been employed as a common method to remove phosphorus from wastewater.^1,11 However, this process can produce large amounts of sludge and has poor removal ability.^4,12 Chemical inhibitors, such as ethylenediaminetetraacetic acid (EDTA), sodium polyphosphate, and silicates, have been also used to reduce the magnesium concentration by forming chelates or less soluble substances with magnesium ions in solution.^1,11 Among these inhibitors, EDTA was shown to be quite efficient, but it usually degrades slowly in the environment.^11,13–15 This has raised environmental concerns about its role in heavy metal mobilization in groundwater.^13,16 For the third approach, inhibition was achieved by the selective binding of inhibitor molecules with some specific crystal faces of struvite, therefore decreasing the struvite growth rate. For example, Wierzbicki et al.¹⁷ found that phosphocitrate preferentially binds to the (101) faces of struvite, which leads to morphological modification or total growth cessation of struvite when sufficient phosphocitrate is used. The fourth strategy is to encourage struvite precipitation by adding MgCl₂ and NaOH into a specific reactor, as the chemical composition and pH of wastewater determine the potential of struvite precipitation.^1,18,19 In this way, the spontaneous precipitation of struvite can be prevented in WWTPs.

Among these approaches, the fourth one has been a research focus in recent years because this crystallization technique is also regarded as an effective way to recover phosphorus fertilizers.^20,21 However, the disadvantages of this technique are apparent: (1) except for the supernatant of digested sludge containing relatively high concentrations of ammonium and phosphate, many wastewater processes cannot fulfill the basic requirement for struvite precipitation.²² (2) The process has never been proven to be commercially profitable due to the supplementation of magnesium salt and sodium hydroxide.^20,22 (3) The incorporation of toxic heavy metals, metalloids (e.g., Cr, Zn, and As) and pathogens into struvite crystals can act potentially as a source of pollution when struvite is used as a fertilizer.^23–28 (4) In sludge liquors, calcium ion levels can be high relative to magnesium.²⁹ Calcium ions can interact with phosphate to form hydroxyapatite or amorphous calcium phosphates, which leads to the inhibition of struvite production and impurity of the recovered product.^20,30 (5) Undesired struvite fine particles are often generated in the crystallization processes due to the high mixing energy (or turbulence) needed to suspend the growing particles, which results in a loss of struvite particles for phosphorus recovery.³¹ (6) The fertilization efficiency of struvite is not superior to other phosphate based compounds such as monocalcium phosphate (Ca(H₂PO₄)₂·H₂O) and dicalcium phosphate (CaHPO₄·2H₂O).³² Therefore, Hao et al.³² pointed out that phosphate recovery should not just focus on struvite.

In view of the harm caused by struvite scale and the disadvantages of the struvite crystallization recovery technique, it is still a challenge to search for effective chemical inhibitors and develop new antiscale techniques. Normally, wastewater tends to be deficient in magnesium ions.²² Therefore, sequestering magnesium ions will be an ideal way to prevent struvite precipitation. Polyaspartic acid (PASP) is a synthetic polyamino acid, which is rich in carboxylic functional groups that can combine with metal ions to form metal–PASP species.^16,33 PASP has been reported to be an effective inhibitor of calcium carbonate, calcium oxalate and calcium phosphate.^34–36 Moreover, PASP is water-soluble, nontoxic, biocompatible, and highly biodegradable, thus it is regarded as the most promising green scale inhibitor in industry.^37,38 However, to the best of our knowledge, no research has been reported on its scale inhibition to struvite. In the present study, we investigate the inhibitory capacity of PASP on struvite in a dynamic environment. The influence of several physicochemical parameters, including pH, mixing energy, reaction time, and the presence of calcium ions on PASP inhibition performance, is systematically investigated. The potential of PASP to dissolve existing struvite scale is also assessed.

2. Materials and methods

2.1. Materials

Magnesium chloride hexahydrate (MgCl₂·6H₂O), ammonium dihydrogen phosphate (NH₄H₂PO₄), ammonium chloride (NH₄Cl), and sodium hydroxide (NaOH) were purchased from Sinopharm Chemical Reagent Co., Ltd, and are of analytical grade. The analytical grade PASP was obtained from Chengdu Ai Keda Chemical Technology Co., Ltd, and its molecular weight is 10 000. Deionized water was used in all experiments.

2.2. Struvite crystallization

All experiments were conducted at room temperature. Synthetic sludge liquor was prepared as described by Doyle et al.³⁹ Magnesium and phosphate solutions with a concentration of 4 mM were used. Because ammonium is always in excess relative to magnesium and phosphate in sludge liquor,²² ammonium chloride was added to produce an excess of ammonium ions. In a typical synthesis procedure, 0.05 g (0.005 mmol) PASP and 0.0407 g (0.2 mmol) MgCl₂·6H₂O were dosed to 40 mL of deionized water in a 50 mL beaker under vigorous stirring to form homogeneous solution A. Then, 0.023 g (0.2 mmol) NH₄H₂PO₄ and 0.0214 g (0.6 mmol) NH₄Cl were dissolved in 10 mL of deionized water to form solution B. Solution B was introduced into solution A under continuous stirring, and a homogeneous synthetic liquor was obtained, with a molar ratio of 1 : 1 : 3 (Mg²⁺ : PO₄³⁻ : NH₄⁺). Subsequently, the pH of the liquor was adjusted to 9.0 by the addition of 0.5 M NaOH. The beaker was then covered with Parafilm to reduce CO₂ interference and NH₃ volatilization, and stirred for 8 h at 360 rpm on a magnetic stirrer. Finally, the product was isolated by centrifugation (1400g for 3 min), washed with absolute alcohol three times, and dried under vacuum at room temperature for 48 h. The same procedures were employed to study the effect of PASP concentration, reaction time, mixing energy (stirring speed), pH, and Mg/Ca ratio on struvite formation. All trials were conducted in triplicate.

2.3. Struvite dissolution

The struvite used for the dissolution was synthesized via homogeneous precipitation by dissolving 0.508 g MgCl₂·6H₂O and 0.288 g NH₄H₂PO₄ in a 50 mL beaker containing 50 mL of deionized water under vigorous stirring. The concentration of Mg²⁺, PO₄³⁻, and NH₄⁺ was 50 mM. The pH of this solution was adjusted to 8.0 using 0.25 M NaOH. The beaker was then covered with Parafilm and kept static. After aging for 4 h, the product was harvested in a similar way as described in Section 2.2. XRD analysis confirmed that the obtained product was pure struvite (data not shown), and SEM observation revealed that the struvite crystals exhibited a coffin-like shape. Dissolution experiments were conducted at room temperature. 0.05 g of PASP was dissolved in 50 mL of deionized water in a 50 mL beaker, and then the pH was adjusted to 8.0 with 0.25 M NaOH. After the addition of 50 mg of struvite powder, the beaker was covered with Parafilm and stirred for 30 min at 360 rpm on a magnetic stirrer. Finally, the remaining precipitate was collected as described in Section 2.2. The same procedures were employed in the PASP concentration-dependent dissolution experiments. All trials were also conducted in triplicate.

2.4. Analytical techniques

X-ray diffraction pattern (XRD) was obtained on an X-ray diffractometer equipped with Cu Kα irradiation (λ = 0.154056 nm, Japan, MapAHF). The morphology and size of the precipitate particles were observed using a field emission scanning electron microscope (FESEM, JEOL JSM-6700F). Energy dispersive X-ray spectroscopy (EDX) analyses of samples coated with Au were obtained with an EDAX detector installed on the same FESEM.

3. Results and discussion

3.1. Effect of PASP concentration on struvite formation

To understand the effect of PASP on struvite crystallization and growth, a series of experiments with PASP concentrations ranging from 0.0 to 0.8 mM were first conducted. The XRD results of the precipitation products confirmed that all the precipitates obtained with and without PASP are orthorhombic struvite with the space group Pmn2₁ (JCPDS file of No. 15-0762), and the representative XRD patterns are shown in Fig. 1a and b. The FESEM results of the struvite precipitates are depicted in Fig. 2. It can be observed from Fig. 2a that a large number of rod-shaped crystals were obtained in the absence of PASP, with a length of ca. 30 μm and a width of ca. 10 μm. When 0.01 mM of PASP was dosed, the resulting product was arrowhead-shaped with a reduced length of ca. 5 μm (Fig. 2b), which is significantly different from the rod-shaped structure. This morphology was retained as the PASP concentration was further increased to 0.05 and 0.1 mM. However, when the PASP concentration reached 0.3 mM, the panoramic FESEM image shows that massive trapezoidal crystals with a length of ca. 7 μm and less rod-like crystals coexisted (Fig. 2c). The further magnified image shows that the “rods” are also trapezoidal (inset in Fig. 2c). As the PASP concentration increased to 0.5 mM, the products occurred in three shapes, which mainly included trapezoids (ca. 7 μm), equilateral triangles (ca. 5 μm), and a few rhomboids (ca. 14 μm) (Fig. 2d). Further increasing the PASP concentration to 0.6 or 0.7 mM did not result in any significant change in morphology. In particular, no precipitate could be formed when 0.8 mM of PASP was used. It appears that PASP significantly impacts not only struvite formation but also its morphogenesis. Fig. 1c quantitatively depicts the effect of PASP on struvite precipitate. Fig. 1c shows that the struvite mass dramatically decreased with the increase in PASP concentration, except for 0.01 mM, which indicates that strong inhibition occurred, and complete inhibition was achieved with the dose of 0.8 mM PASP. Doyle et al.¹¹ tested the effectiveness of EDTA on preventing struvite precipitation. According to their study, as much as 8 mM of EDTA is needed to achieve the complete inhibition of struvite precipitation. In this regard, PASP is much more effective than EDTA. Moreover, the size of struvite crystals obtained in the presence of PASP was markedly reduced. This will make struvite easier to be washed away and harder to scale.

Fig. 1 Typical XRD patterns of the precipitates from the artificial liquor dosed with 0 (a) and 0.5 (b) mM PASP. (c) Effect of PASP concentration on mass of struvite precipitation. All the precipitation runs were performed at pH_i 9.0 for 8 h.

Fig. 2 FESEM images of the 8 h struvite crystals from the artificial liquor dosed with 0 (a), 0.01 (b), 0.3 (c), and 0.5 (d) mM of PASP. Inset: the magnification of the rod-like crystals appeared in (c).

It is well known that PASP is a good chelating agent, and it can coordinate with a variety of metal cations to form complex species.^16,33 Therefore, its coordination with Mg²⁺ can be expressed as follows:

Mg²⁺ + nPASP ⇆ Mg(PASP)_n (1)

Mg²⁺ + NH₄⁺ + H₂PO₄⁻ + 6H₂O ⇆ MgNH₄PO₄·6H₂O↓ + 2H⁺ (2)

where the coordination and precipitation reactions (1) and (2) compete with each other, and precipitation will predominate with an increased pH based on reaction (2). With the precipitation of struvite, free Mg²⁺ ions decrease and the coordination equilibrium shifts towards the disassociation of Mg–PASP complexes, which leads to the increase of free PASP. The released PASP molecules can selectively bind to the crystal faces of struvite, and thus affect the crystal growth habit (e.g., Fig. 2). Therefore, PASP molecules play dual roles during scale inhibition: morphological modification and precipitation inhibition of struvite. In fact, similar arrowhead-shaped struvite has been observed in the presence of PASP in our previous biomimetic mineralization experiments,⁴⁰ and the selective recognition and binding of PASP onto the {010} and {101} faces of struvite crystals were believed to be responsible for the formation of this specific morphology. These faces have a high density of magnesium cations, and therefore provide a positively charged environment for the preferential adsorption and binding of PASP molecules that are rich in negatively charged carboxyl side chains.^17,40,41 The PASP molecules that bind to these faces decrease their growth rate and this leads to enhanced expression of these faces, thus modifying crystal morphology. Similarly, the trapezoids and equilateral triangles observed in our case may also result from the preferential adsorption and binding of PASP onto some specific faces of struvite crystals. As for the increase of struvite precipitate at 0.01 mM of PASP, this is possible because less PASP molecules in the liquor cannot cause a significant decrease in free Mg²⁺ by complexation with Mg²⁺ ions. In contrast, a low PASP concentration can induce Mg²⁺ and NH₄⁺ aggregation and accumulation around them due to their carboxylic groups, and promote struvite nucleation. As a result, less PASP molecules in the liquor act as a nucleation template, and thus facilitate the precipitation of struvite.⁴² A similar effect has also been reported by Elhadj et al.⁴³ in a system of PASP and calcite. However, with the increase of PASP concentration, the coordination between PASP and Mg²⁺ predominates the template function of PASP. Therefore, high concentrations of PASP can effectively block struvite precipitation and growth; even completely avoid the formation of struvite precipitate.

3.2. Effect of reaction conditions on PASP inhibition performance

The influence of several key parameters, including reaction time, mixing energy, pH, and calcium ions on the scale inhibition performance of PASP was examined. 0.5 mM PASP was dosed for each run to guarantee an effective and incomplete inhibition of struvite formation, and other conditions were the same as the PASP concentration-dependent experiments.

3.2.1. Reaction time. If PASP is to be used as an inhibitor to prevent struvite formation in WWTPs, its inhibitory capacity should last long enough to allow wastewater to pass through the entire pipe before precipitation. Therefore, experiments with stirring times of 8, 16, 24, and 48 h were carried out. Precipitates were always formed at different time intervals due to the limited dose of PASP (0.5 mM). The XRD analyses confirmed that these precipitates were also struvite (e.g., Fig. 1b). The FESEM results showed that the morphology and size of struvite obtained after 16, 24, and 48 h were the same as the 8 hour product (e.g., Fig. 2d). It appears that different stirring times have no effect on precipitate phase and morphology. Fig. 3a shows a plot of precipitate mass versus stirring time. Although a slight increase in precipitate mass was observed from 8 to 16 h, the precipitate mass remained almost constant after 16 h of stirring, which indicates that the magnesium–PASP species are stable. Therefore, PASP inhibition performance is effective and sustainable in a longer period of time, thus ensuring that wastewater can be treated with less struvite precipitation.

Fig. 3 Mass of the precipitates obtained under different precipitation conditions: (a) different time intervals; (b) different stirring speeds; (c) different pH_i values; (d) different Mg/Ca ratios.

3.2.2. Mixing energy. It is commonly observed that struvite preferentially accumulates in specific locations of a treatment or conveyance system with high mixing energy, such as pipe elbows, pumps, and mixers, rather than uniform deposition.⁴⁴ The mechanisms responsible for the preferential accumulation of struvite are associated with the mixing energy (or turbulence). In the areas of high mixing energy, CO₂ liberation is enhanced. This can increase the pH of the solution, and therefore favor the formation of struvite crystals.²⁰ More importantly, struvite growth rate was found to be transport-limited.⁴⁴ Thus, high mixing energy will enhance the mass transfer of solute to the crystals and facilitate struvite crystallization and growth.^44,45 In our experiments, we tested the effect of mixing energy on PASP inhibition performance with stirring speeds ranging from 0 to 960 rpm (i.e., 0, 120, 360, 600, and 960 rpm). After 8 h of reaction, no product was obtained in the static experiment (0 rpm) and the solution was always clear. In contrast, struvite crystals with similar morphology and size were harvested under magnetic stirring ranging from 120 to 960 rpm (e.g., Fig. 1a and 2d), which reveals that stirring favors the nucleation and growth of struvite. However, no marked changes in precipitate mass were observed with the increase in stirring speed (Fig. 3b), which also further indicates that in the presence of PASP the high mixing energy cannot transport enough free Mg²⁺ to the growth fronts of struvite crystals, and therefore accelerates crystal growth. In other words, the high mixing energy cannot weaken the inhibitory capacity of PASP, and PASP can be potentially applied for controlling struvite scaling at different locations, especially high-mixing environments.

3.2.3. Initial pH. pH can affect the dissolution and supersaturation, as well as morphology, particle size, and purity of struvite.^20,46 Wastewater may have different pH values depending on the type and source, and pH cannot be always controlled or maintained throughout the struvite crystallization process.⁴⁶ Therefore, the effect of initial pH (pH_i) on PASP performance was investigated. Herein, pH_i ranged from 8 to 11 (i.e., 8, 9, 10, 11). Control experiments without PASP were also conducted. In the absence of PASP, the collected precipitates were all rod-shaped struvite at pH_i 8–10 (Fig. 1a, 2a, S1a, b and S2a†). At pH_i 11, the precipitate consists of rod-shaped struvite and plate-like cattiite [Mg₃(PO4)₂·22H₂O], which were confirmed by XRD and EDX analyses (Fig. S1c, S2b, S3a and b†). In the presence of PASP, no precipitate was obtained at pH_i 8, but pure struvite was formed at pH_i 9–11 (Fig. 1a, S1d and e†), and these struvite crystals became slimmer with pH_i (Fig. 2d, S2c and d†). The variation trend of struvite shape agrees well with the results reported by Ma et al.⁴⁶ This is because a high precipitation rate and changes in aqueous speciation decrease the concentrations of Mg²⁺(aq), NH₄⁺(aq), and PO₄³⁻(aq), which results in the limiting of crystal growth at higher pH_i.⁴⁶Fig. 3c shows the effect of pH_i on the mass of precipitate products in the presence of 0.0 and 0.5 mM PASP. It can be observed that the variation trends are similar under these two conditions. Specifically, mass increased with pH_i, and the highest was at pH_i 10. This is consistent with the previous observations by Ma and Rouff.²⁴ The increase in yield with pH_i can be attributed to an increase in the activity of PO₄³⁻(aq) because an increase in pH promotes an equilibrium shift from H₂PO₄⁻ and HPO₄²⁻ to PO₄³⁻ species. However, at pH_i 11, this effect is neutralized by the reduced activity of NH₄⁺(aq) due to the increased formation of NH₃ species, and by the hydrolysis of Mg²⁺(aq) to MgOH⁺(aq).²⁴ It is notable that in the presence of PASP, the precipitation of struvite was significantly inhibited at pH_i from 8 to 11 compared with the controls (Fig. 3c). Therefore, PASP can inhibit struvite growth over a large pH_i range, and an additional dose of PASP can achieve the same inhibitory efficiency at higher pH_i.

3.2.4. Mg/Ca ratio. In sludge liquors, calcium levels can be high relative to magnesium.²⁹ These calcium ions can interact with phosphate or carbonate ions to form additional mineral precipitates such as hydroxylapatite and calcite.²⁰ In this context, struvite formation can be inhibited if the supply of phosphate is limited. As a chelating agent, PASP can form stable complexes with calcium and magnesium ions in solution.^16,33 The presence of calcium ions will unavoidably reduce the inhibitory efficiency of PASP on struvite. Therefore, inhibitory experiments at Mg/Ca ratios of 1 : 0, 2 : 1, 1 : 1, and 1 : 2 were performed. Similarly, the precipitates were examined using the FESEM and XRD techniques. In the absence of PASP, the precipitates exhibited different mineralogical characteristics over the range of Mg/Ca ratios. Without the addition of Ca, pure rod-shaped struvite was obtained (Fig. 1a and 2a). When the Mg/Ca ratio was 2 : 1, rod-shaped crystals coated with nanoparticles and irregular aggregates of nanoparticles were formed (Fig. S4a†). The XRD pattern, despite the slight background noise, could still be well indexed as struvite (Fig. 4a). Therefore, the rod-like crystals can be safely assigned to struvite, whereas the nanoparticles and their aggregates may be an amorphous precipitate. Further increase in the Ca concentration (Mg/Ca = 1 : 1) led to the enhanced output of irregular aggregates with a few rod-like crystals (Fig. S4b†), and the much more noise can be observed from the XRD pattern (Fig. 4b), which is indicative of the formation of a large amount of amorphous matter. When the Mg/Ca ratio reached 1 : 2, only huge tabular aggregates were harvested (Fig. S4c†), and XRD analysis confirmed that the aggregates exhibited an amorphous feature (Fig. 4c), which indicates that Ca²⁺ present in the liquor facilitates the formation of amorphous precipitate. To further understand the chemical nature of the amorphous aggregates, EDX analyses were also conducted (Fig. S3c–g†). Combined with the FESEM images and XRD patterns, the aggregates were identified as amorphous calcium phosphate. le Corre et al.²⁹ also obtained amorphous calcium phosphate when they studied the impact of calcium on struvite growth. We also found that the addition of Ca inhibited struvite growth but increased the total yield of the precipitates, i.e., the formation of other scales (Fig. 3d). However, in the presence of 0.5 mM PASP, only trapezoidal and rod-like struvite was obtained at different Mg/Ca ratios, and no calcium precipitate was harvested (Fig. S4d–f and 4d–f). This indicated that calcium ions have priority to complex with PASP, which results in the decrease of PASP inhibition performance on struvite formation. Therefore, a slight increase in struvite mass was observed with an increase in calcium concentration (Fig. 3d). However, the mass of struvite obtained in the presence of PASP was still very low at different Mg/Ca ratios compared with the experiments without PASP, which means that the effect of calcium on PASP inhibition performance was limited and PASP can still exert significant scale inhibition even in the presence of calcium.

Fig. 4 XRD patterns of the precipitates at Mg/Ca ratio 2 : 1 (a), 1 : 1 (b), 1 : 2 (c) dosed with 0 mM PASP, and at Mg/Ca ratio 2 : 1 (d), 1 : 1 (e), 1 : 2 (f) dosed with 0.5 mM PASP.

3.3. Effect of PASP concentration on struvite dissolution

Except for the inhibition of scale formation, the removal of existing struvite scale is another important effort in WWTPs. Investigators have tried to find out effective and environmentally safe chelating agents to provide a feasible alternative to acid dissolution, which can cause corrosion in metallic process equipment and piping systems.⁴⁷ Previous studies demonstrate that PASP can promote the dissolution of calcium oxalate and calcium phosphate due to its ability to chelate metal ions in solution.^47–49 Therefore, the chelation of PASP with magnesium may also be a driving force for struvite dissolution. Herein, the dependence of struvite dissolution on PASP concentration was examined at pH 8. Fig. 5 presents FESEM images of the products before and after dissolution with different concentrations of PASP. The pristine struvite crystals have a coffin-like shape with a length of ca. 30 μm and a width of ca. 10 μm, and a number of tiny pits and crevices can be observed on their lateral sides (Fig. 5a). After dispersion in deionized water for 30 minutes, some of the tiny pits or crevices were enlarged, but the basic configuration was retained, which indicates that limited dissolution occurred (Fig. 5b). Although struvite solubility is low in water, it can be 18 mg/100 mL at 25 °C.²⁰ Specifically, as much as 9 mg of struvite will be dissolved if dissolution equilibrium is achieved in our case (50 mg struvite + 50 mL deionized water). Our dissolution experiment revealed that 8.3 mg (16.6%) of struvite was dissolved in deionized water, which approaches the equilibrium dissolution value (Fig. S5†). When 0.1 mM of PASP was dosed, the struvite crystals became thinner, and contained many grooves and deep carvings (Fig. 5c). Noticeably, a couple of enlarged corrosion pits could be observed on their surfaces. Some struvite crystals had a decreased width and approximately 28.1 mg (56.2%) of struvite was dissolved (Fig. S5†). Therefore, the dose of PASP can effectively enhance struvite dissolution. When the PASP concentration was 0.5 mM, more remarkable dissolution occurred, which resulted in a much thinner morphology, i.e., dumbbell-shaped (Fig. 5d). Consistently, over 44 mg (88%) of struvite was dissolved in this case (Fig. S5†). Therefore, the effectiveness of PASP to promote struvite dissolution is proportional to its concentration. Wu and Grant¹⁶ reported that all the carboxyl groups of PASP will be ionized above pH 6, which leads to the strongest complexation with magnesium. The complexation reduced the amount of free magnesium ions and therefore broke down the precipitation–dissolution equilibrium, thus resulting in the enhancement of struvite dissolution (eqn (2)). With increasing PASP concentrations, more complexation will occur, and thus further promote dissolution. Therefore, PASP can be an effective cleaning agent for existing struvite scale.

Fig. 5 FESEM images of struvite crystals before (a) and after 30 min of dissolution dosed with 0 (b), 0.1 (c), and 0.5 (d) mM of PASP.

4. Conclusions

In summary, precipitation experiments dosed with PASP showed that PASP can not only effectively inhibit the formation and growth of struvite, but also significantly change struvite morphology, which leads to the evolution from rod-shape to arrowhead-shape, triangles, or trapezoids. Moreover, several key parameters that possibly affect PASP inhibition performance were also tested, and the results revealed that PASP can still exert strong antiscaling on struvite even in a long running time and a large range of mixing energies. However, its inhibition potency is pH dependant, and decreases with pH_i. The presence of calcium ions slightly reduced the inhibition potency of PASP while the dose of PASP prevented the formation of amorphous calcium phosphate. Dissolution experiments dosed with PASP showed that PASP can promote the dissolution of preformed struvite and its effectiveness increases with concentration. It appears that PASP has a strong ability to not only effectively inhibit the formation and growth of struvite, but also facilitate struvite dissolution. Therefore, PASP can serve as an environmentally friendly scale inhibitor and scale cleaning agent.

Acknowledgements

This study was partially supported by the Chinese Ministry of Science and Technology (No. 2014CB846003), the Natural Science Foundation of China (No. 41372053), and the Specialized Research Fund for the Doctoral Program of Higher Education (No. 20133402130007).

References

1. J. D. Doyle and S. A. Parsons, Water Res., 2002, 36, 3925–3940.
2. F. Abbona and R. Boistelle, J. Cryst. Growth, 1979, 46, 339–354.
3. C. Beal, T. Gardner, W. Ahmed, C. Walton and D. Hamlyn-Harris, in Proceedings of AWA Conference, 2007, pp. 1–8.
4. D. Mamais, P. A. Pitt, Y. W. Cheng, J. Loiacono and D. Jenkins, Water Environ. Res., 1994, 66, 912–918.
5. C. M. Mehta and D. J. Batstone, Water Res., 2013, 47, 2890–2900.
6. X. Mohajit, K. K. Bhattarai, E. P. Taiganides and B. C. Yap, Biol. Wastes, 1989, 30, 133–147.
7. J. D. Doyle, K. Oldring, J. Churchley and S. A. Parsons, Water Res., 2002, 36, 3971–3978.
8. S. Williams, Environ. Technol., 1999, 20, 743–747.
9. J. Borgerding, J.-Water Pollut. Control Fed., 1972, 44, 813–819.
10. S. A. Parsons and J. D. Doyle, Water Sci. Technol., 2004, 49, 177–182.
11. J. D. Doyle, K. Oldring, J. Churchley, C. Price and S. A. Parsons, J. Environ. Eng., 2003, 129, 419–426.
12. K. Fytianos, E. Voudrias and N. Raikos, Environ. Pollut., 1998, 101, 123–130.
13. J. L. Means, T. Kucak and D. A. Crerar, Environ. Pollut., Ser. B, 1980, 1, 45–60.
14. V. Sykora, P. Pitter, I. Bittnerova and T. Lederer, Water Res., 2001, 35, 2010–2016.
15. J. Prywer and M. Olszynski, J. Cryst. Growth, 2013, 375, 108–114.
16. Y. T. Wu and C. Grant, Langmuir, 2002, 18, 6813–6820.
17. A. Wierzbicki, J. D. Sallis, E. D. Stevens, M. Smith and C. S. Sikes, Calcif. Tissue Int., 1997, 61, 216–222.
18. I. Stratful, M. D. Scrimshaw and J. N. Lester, Water Res., 2001, 35, 4191–4199.
19. Y. Ueno and M. Fujii, Environ. Technol., 2001, 22, 1373–1381.
20. K. S. le Corre, E. Valsami-Jones, P. Hobbs and S. A. Parsons, Crit. Rev. Environ. Sci. Technol., 2009, 39, 433–477.
21. C. C. Wang, X. D. Hao, G. S. Guo and M. C. M. van Loosdrecht, Chem. Eng. J., 2010, 159, 280–283.
22. L. E. de-Bashan and Y. Bashan, Water Res., 2004, 38, 4222–4246.
23. L. Decrey, K. M. Udert, E. Tilley, B. M. Pecson and T. Kohn, Water Res., 2011, 45, 4960–4972.
24. N. Ma and A. A. Rouff, Environ. Sci. Technol., 2012, 46, 8791–8798.
25. A. A. Rouff, Environ. Sci. Technol., 2012, 46, 12493–12501.
26. J. R. Lin, N. Chen and Y. M. Pan, Environ. Sci. Technol., 2013, 47, 12728–12735.
27. J. R. Lin, N. Chen and Y. M. Pan, Environ. Sci. Technol., 2014, 48, 6938–6946.
28. A. A. Rouff and K. M. Juarez, Environ. Sci. Technol., 2014, 48, 6342–6349.
29. K. S. le Corre, E. Valsami-Jones, P. Hobbs and S. A. Parsons, J. Cryst. Growth, 2005, 283, 514–522.
30. X. D. Hao, C. C. Wang, L. Lan and M. C. M. van Loosdrecht, Water Sci. Technol., 2008, 58, 1687–1692.
31. K. S. le Corre, E. Valsami-Jones, P. Hobbs, B. Jefferson and S. A. Parsons, Water Res., 2007, 41, 419–425.
32. X. D. Hao, C. C. Wang, M. C. M. van Loosdrecht and Y. S. Hu, Environ. Sci. Technol., 2013, 47, 4965–4966.
33. S. D. Jiang, Q. Z. Yao, G. T. Zhou and S. Q. Fu, J. Phys. Chem. C, 2012, 116, 4484–4492.
34. X. X. Sheng, M. D. Ward and J. A. Wesson, J. Am. Chem. Soc., 2003, 125, 2854–2855.
35. A. Bigi, B. Bracci, S. Panzavolta, M. Iliescu, M. Plouet-Richard, J. Werckmann and D. Cam, Cryst. Growth Des., 2004, 4, 141–146.
36. B. Njegic-Dzakula, L. Brecevic, G. Falini and D. Kralj, Cryst. Growth Des., 2009, 9, 2425–2434.
37. D. Hasson, H. Shemer and A. Sher, Ind. Eng. Chem. Res., 2011, 50, 7601–7607.
38. K. L. Dziak and O. Akkus, J. Bone Miner. Metab., 2008, 26, 569–575.
39. J. D. Doyle, R. Philp, J. Churchley and S. A. Parsons, Process Saf. Environ. Prot., 2000, 78, 480–488.
40. H. Li, Q. Z. Yao, Y. Y. Wang, Y. L. Li and G. T. Zhou, Sci. Rep., 2015, 5, 7718.
41. Z. Romanowski, P. Kempisty, J. Prywer, S. Krukowski and A. Torzewska, J. Phys. Chem. A, 2010, 114, 7800–7808.
42. J. Roque, J. Molera, M. Vendrell-Saz and N. Salvado, J. Cryst. Growth, 2004, 262, 543–553.
43. S. Elhadj, E. A. Salter, A. Wierzbicki, J. J. de Yoreo, N. Han and P. M. Dove, Cryst. Growth Des., 2006, 6, 197–201.
44. K. N. Ohlinger, T. M. Young and E. D. Schroeder, J. Environ. Eng., 1999, 125, 730–737.
45. D. Kim, J. Kim, H. D. Ryu and S. I. Lee, Bioresour. Technol., 2009, 100, 74–78.
46. N. Ma, A. A. Rouff and B. L. Phillips, ACS Sustainable Chem. Eng., 2014, 2, 816–822.
47. F. Littlejohn, C. S. Grant, Y. L. Wong and A. E. Saez, Ind. Eng. Chem. Res., 2002, 41, 4576–4584.
48. F. Poumier, P. Schaad, Y. Haikel, J. C. Voegel and P. Gramain, J. Biomed. Mater. Res., 1999, 45, 92–99.
49. S. W. Guo, M. D. Ward and J. A. Wesson, Langmuir, 2002, 18, 4284–4291.

---

Footnote

† Electronic supplementary information (ESI) available: The characterizations of the samples precipitated at different pH_i and different Mg/Ca ratios, including XRD patterns, FESEM images, and EDX spectra. Plot of remaining mass in the dissolution experiments. See DOI: 10.1039/c5ra17149k

---