*3.2. Effect of Phosphorus Concentration*

To study the effect of the initial phosphorus concentration on struvite crystallization, we performed a series of experiments at a constant stirring rate of 100 rpm, a solution volume of 600 mL, and an initial phosphorus concentration ranging from 200 to 800 mg·L−<sup>1</sup> (6.45 to 25.8 mM, respectively). The FTIR spectra of the precipitates obtained at different initial phosphorus concentrations are given in Figure 6. These spectra were comparable to the struvite spectrum. Therefore, the variation in the initial phosphorus concentration up

to 800 mg·L<sup>−</sup>1, at a constant stirring rate of 100 rpm and a solution volume of 600 mL, did not affect the nature of the precipitated phase.

**Figure 6.** FTIR spectra of the precipitates obtained for different initial phosphorus concentrations, a stirring rate of 100 rpm, and a solution volume of 600 mL.

Changes in pH and phosphorus concentration over time for different initial phosphorus concentrations, a stirring rate of 100 rpm, and a solution volume of 600 mL (not presented herein) showed the same evolution trends as those obtained for different stirring rates, an initial phosphorus concentration of 600 mg·L−1, and a solution volume of 600 mL presented in Figure 4. That is, the pH and the phosphorus concentration decreased over time, and then from a certain time remained practically constant, which indicated the end of struvite precipitation. The end of precipitation time (tf), the initial precipitation rate (Vi), and the phosphorus removal efficiency *R* (%) obtained for the different initial phosphorus concentrations are given in Table 3. The end of precipitation time decreased with increasing the initial phosphorus concentration, while the initial precipitation rate increased. For example, tf decreased from 50 to 20 min and Vi increased from 2.33 to 12.55 mmol·L−1·min−<sup>1</sup> when the phosphorus concentration increased from 300 to 800 mg·L−1, respectively (Table 3). The phosphorus removal efficiency increased from ~87% to ~99% when the initial phosphorus concentration increased from 300 to 800 mg·L<sup>−</sup>1, respectively. Therefore, the increase in the initial phosphorus concentration, at a fixed stirring rate of 100 rpm and a solution volume of 600 mL, improved phosphorus recovery, and more struvite crystals were precipitated at a shorter experiment time.


**Table 3.** End of precipitation time (tf), the initial precipitation rate (Vi), and the phosphorus removal efficiency (*R*%) obtained for different initial phosphorus concentrations ([P]0), a stirring rate of 100 rpm and a solution volume of 600 mL.

#### *3.3. Particle Size*

For an initial phosphorus concentration of 200 mg·L−1, a solution volume of 600 mL and varied stirring rates, the particle size distribution curves showed a polymodal distribution, i.e., two maxima were observed for a stirring rate of 100 rpm, and three maxima were observed for a stirring rate of 700 rpm (Figure 7a). For both stirring rates, the first maximum was located at ~18 μm and the second one was located at ~100 μm. The third maximum

observed at a stirring rate of 700 rpm appeared at ~875 μm. Therefore, the increase in the stirring rate led to the formation of large struvite crystals. This can be explained by a greater transfer of the struvite constituent ions from the liquid phase (solution) to the solid phase (the germs formed) for higher stirring. At a stirring rate of 700 rpm and an initial phosphorus concentration of 200 mg·L−1, when the volume of the solution increased, the particle size distribution curves showed a polymodal distribution and a decrease in struvite particle size was observed (Figure 7b). For example, for a volume of 800 mL, the size distribution was trimodal and the maxima observed were located at ~2, ~18, and ~100 μm. For higher volumes of 1200 and 1400 mL, the size distribution became bimodal, and the maxima were located at ~18 and ~100 μm.

**Figure 7.** Particle size distributions of struvite crystals for (**a**) selected stirring rates at an initial phosphorus concentration of 200 mg·L−<sup>1</sup> and a solution volume of 600 mL, (**b**) different solution volumes at a stirring rate of 700 rpm, and (**c**) different initial phosphorus concentrations at a stirring rate of 100 rpm and a solution volume of 600 mL.

The particle size distributions of struvite crystals for phosphorus concentrations ranged from 200 to 700 mg·L−1, a stirring rate of 100 rpm and a solution volume of 600 mL showed a bimodal distribution (Figure 7c). For a phosphorus concentration of 800 mg·L−1, the size distribution became trimodal. For 200 and 300 mg·L−1, the first maximum was observed at ~10 μm and the second one at ~100 μm. For initial phosphorus concentrations between 400 and 700 mg·L−1, the first maximum was observed at ~30 <sup>μ</sup><sup>m</sup> and the second one at ~100 <sup>μ</sup>m. For a phosphorus concentration of 800 mg·L−1, besides the two maxima at 30 and 100 μm, a third maximum was observed at ~400 μm. Therefore, the increase in initial phosphorus concentration led to an increase in struvite particle size.

In summary, at an initial phosphorus concentration of 200 mg·L−<sup>1</sup> and a constant volume of 600 mL, increasing the stirring rate up to 500 rpm accelerated the precipitation of struvite, improved the phosphorus removal efficiency, and obtained larger struvite crystals. According to the results of Natsi et al. [7] and considering the dependence of phosphorus recovery through struvite crystallization on the solution turbulence depicted in the present work, the predominant mechanism of struvite crystallization can be surfacediffusion controlled. Indeed, the increase in the stirring rate from 100 to 500 rpm led to significant turbulence of the solution, which accelerated the diffusion of the constituent ions to the germs formed in the solution, and therefore an acceleration of struvite crystalline growth occurred. As a result, the particle size increased. This is of utmost importance from an experimental point of view since large struvite crystals formed in the solution facilitate its separation from the liquid phase. In addition, struvite shows slow-release properties [40] and a large surface area to volume ratio accelerates the release of ammonium and phosphorus from struvite used as fertilizer [41]. Increasing the stirring rate from 500 to 700 rpm to maintain more precipitates in suspension for a larger effective precipitation area resulted, however, in a decrease in phosphorus removal, and smaller struvite crystals were obtained. Indeed, when the stirring rate significantly increased, the mixing energy increased as well, the flow became turbulent, and the liquid shear stress was higher. This

affected the transfer of struvite constituent ions from the liquid phase to the solid phase and inhibited their reaction with the crystal surfaces, limiting struvite crystal growth. By increasing the volume of the solution at a high stirring rate of 700 rpm, the turbulence of the solution decreased, and therefore, the diffusion of the constituent ions became less important, which led to smaller particles. Increasing the initial phosphorus concentration overcame the limiting effect of turbulence and resulted in an increase in struvite particle size.

In real wastewater, the concentration of total nitrogen can vary from ~50 mg·L<sup>−</sup>1, i.e., in municipal wastewater with a low C/N ratio [42], to ~2700 mg·L−<sup>1</sup> in swine wastewater, for example [16]. The concentration of phosphorus and magnesium can vary from a few tens of mg·L−<sup>1</sup> to concentrations near 900 mg·L−<sup>1</sup> [14]. The constituent ion concentrations used in the present work fitted well with these ranges of concentrations. Real wastewater may also contain calcium (~20–60 mg·L−<sup>1</sup> [13,14,16] ), chlorine (~90 mg·L−<sup>1</sup> [16]), and low concentrations of heavy metals such as Cu (~2 mg·L−1), Zn (~6 mg·L−1), Cd (~0.5 mg·L−1), Pb (~0.8 mg·L−1), Mn (~1 mg·L−1) [16], and Fe (~5 mg·L−1) [43]. Note that the concentrations are given as an indication. These foreign ions can adsorb on struvite surfaces, incorporate in the struvite lattice, or precipitate as separate phases that disturb struvite growth in all cases. For example, calcium precipitated in treated wastewater as calcium phosphate salts [44,45], which affected struvite crystallization. Copper and iron precipitated in the form of hydroxides during struvite crystallization [43], and zinc precipitated as Zn-PO4 at low Zn concentration and Zn-OH at high Zn concentration [46]. Depending on the added concentration, lead can be adsorbed on struvite surfaces ([Pb] < 1 mg·L−1) or precipitated as Pb hydroxide and Pb phosphate ([Pb] between 10 and 100 mg·L−1) [47]. Like lead, the presence of nickel in the wastewater competed with struvite crystallization, i.e., at low concentrations (<1 mg·L−1), nickel formed Ni-OH and Ni-PO4 on the struvite surface and precipitated separately as amorphous Ni-struvite, Ni hydroxide, and Ni phosphate at higher concentrations between 10 and 100 mg·L−<sup>1</sup> [20]. Additionally, we showed in the present work that under a constant stirring rate, in the range between 100 and 500 rpm, the increase in the solution volume from 600 to 1200 mL did not significantly affect the struvite crystal growth, and for a higher stirring rate of 700 rpm, the volume of the solution should be reduced to obtain high amounts of struvite. However, experiments in the present work were conducted on free foreign ion solutions with relatively low volumes that did not exceed 1.4 L. Obviously, the volumes of wastewater treated in industrial processes are significantly larger than those used in the present work. For those reasons, further work is needed to investigate struvite crystallization from real wastewater at laboratory and industrial scales.

#### **4. Conclusions**

Phosphorus and ammonium can be recovered in the presence of magnesium through struvite (MgNH4PO4·6H2O) crystallization. Struvite is recognized as an effective fertilizer. In the present work, the crystallization of struvite by magnetic stirring was investigated at different initial phosphorus concentrations, solution volumes, and stirring rates. The crystals obtained were characterized by XRD, FTIR spectroscopy, and SEM. For all experiments, struvite was the only solid observed. It was shown that:


**Author Contributions:** A.K. was involved in the methodology, experimental investigation, supervision, and writing and editing of the work. S.A. was involved in the experimental investigation. I.S. was involved in ensuring resources, data curation, and formal analysis of the work. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Deanship of Scientific Research at King Khalid University, Saudi Arabia, grant number GRP/87/44.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not available.

**Acknowledgments:** The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University, Saudi Arabia, for funding this work through General Research Project under grant number GRP/87/44.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
