Chemical precipitation granular sludge (CPGS) formation for copper removal from wastewater

Abstract

Solid–liquid separation and the settleability of sludge are always issues in copper wastewater treatment by chemical precipitation. To improve the settling performance of chemical precipitation sludge, controlled double-jet precipitation (CDJP) was used to treat wastewater by generating chemical precipitation granular sludge (CPGS). Several parameters such as the pH, copper oxide (CuO) seed dosage, stirring rate, adding rate and polyacrylamide dosage affecting the formation of the CPGS and wastewater treatment were investigated. Results show that the Cu(II) removal efficiency is over 99% and the settling velocity of the CPGS can reach up to 8.1 cm s⁻¹ under optimized conditions, i.e., pH of 11, seed type of common CuO, seed dosage of 2.5 g L⁻¹, stirring rate of 600 rpm, adding rate of 0.5 mL min⁻¹ and PAM dosage of 20 g L⁻¹. This settling velocity was about 18 times higher than that of chemical flocculent sludge. More importantly, the CPGS formation mechanism was explored. A three-step formation mechanism was proposed involving fresh precipitate growth on the seed surface, flocculation, and rearrangement with stirring.

---

1. Introduction

Copper is a kind of common heavy metal that serves as a micronutrient element. It plays an important role in bone formation together with certain proteins and enzymes.¹ However, if Cu cannot be maintained at an appropriate physiological concentration, it causes serious syndrome diseases, such as osteoporosis disease, Wilson's disease, and Alzheimer's disease.² With the widespread use of copper materials, there are many actual and potential sources of copper pollution, such as mining, metallurgy, plating and printing circuits.^3,4 In particular, large quantities of Cu-containing wastewater are generated due to the disposal of industrial waste. Copper in wastewater exhibits high mobility and can easily diffuse into the environment by forming complex compounds. Therefore, copper removal from water and wastewater has a significant importance for curbing its toxicity towards humans, animals and the environment.

A large number of technologies have been employed so far to ensure the environmental safety against Cu(II) in the industrial effluents, including chemical precipitation,⁵ membrane filtration,⁶ ion exchange,⁷ reverses osmosis,⁸ coagulation,⁹ electrolysis,¹⁰ and sorption.^11–14 Among these methods, hydroxide precipitation is one of the most widespread technologies due to its simplicity, low cost and easy operation.¹⁵ However, hydroxide precipitation also has some limitations. It generates large volumes of low density sludge, which can cause dewatering and disposal problems.¹⁶ There is no doubt increases the operation cost of sludge post-treatment process. Various methods have been developed to improve settling performance of chemical precipitation sludge, such as adding seed materials,¹⁷ pressurised electro-osmotic,¹⁸ magnetic separation,^19,20 returned sludge method,^21,22 sonication,^23,24 coagulation and flocculation.^25,26 However, the above single and traditional approach cannot substantially change the characteristics of flocculent chemical precipitation sludge. Hence, there is an urgent need to develop a new strategy to improve the settleability of chemical precipitation sludge.

Recently, our research group²⁷ proposed a novel and efficient strategy to improve the settling performance of chemical precipitation sludge by formation granular sludge for zinc containing wastewater treatment. The formed sludge possessed a dense structure, large size, and regular spherical morphology. Its settling velocity can reach up to 3.0 cm s⁻¹, which is similar with the biological granular sludge (3.5 ± 0.4 cm s⁻¹)²⁸ and much higher than chemical precipitation sludge (<1 cm s⁻¹).²⁹ However, systematic studies for the applicability of this strategy in copper containing wastewater treatment have not been investigated.

In this study, the CPGS formation strategy was used to treat Cu-containing wastewater. The effects of the pH, CuO seed dosage, stirring rate, flow rate of reactants, and polyacrylamide (PAM) dosage on Cu removal efficiency and sludge settling performance were explored. In addition, the sludge formation mechanism was attempted to shed some light on the precipitation process.

2. Materials and methods

2.1 Materials

All chemicals were analytical grade and used without further purification. Simulated wastewater containing 1000 mg L⁻¹ Cu(II) was prepared by dissolving copper(II) sulfate pentahydrate (CuSO₄·5H₂O) in deionized water. Sodium hydroxide (NaOH) was purchased from Sinopharm Chemical Reagent Co., Ltd. Common copper oxide (CuO) seed was purchased from Sinopharm Chemical Reagent Co., Ltd., which is irregular shape. Nano-CuO seed was purchased from Aladdin Reagent Co. Ltd., which is 40 nm and spherical. A nonionic PAM was purchased from Kemiou Chemical Reagent Co., Ltd. and used as flocculant. PAM solutions (1000 mg L⁻¹) were prepared by completely dissolving the powdered PAM in deionized water, followed by ageing for 24 h prior to use.

2.2 Procedures

100 mL of distilled water containing a certain amount of CuO powder was added into the reactor. Next, simulated wastewater and 0.03 M NaOH solution were simultaneously introduced by peristaltic pumps into the controlled double-jet precipitation (CDJP) reactor at a constant adding rate and stirring rate. The total injection time was set to 60 min at room temperature. The pH was kept in 11.0 ± 0.2 by dripping 0.03 M NaOH solution and was constantly monitored using a pH meter. Then a certain dosage of PAM was added with a rapid stirring (500 rpm) for 30 s, followed by a slow stirring (100 rpm) for 15 min to allow CPGS growth. Finally, a quiescent sedimentation phase was maintained for 30 min.

2.3 Analysis methods

After sedimentation, residual Cu(II) concentration in supernatant was measured by a WFX-200 atomic absorption spectrometry (Rayleigh, Beijing, China). To prevent clogging of the piping by precipitation in the supernatant, it was filtered to eliminate as-prepared particles and then used nitric acid to digeste supernatant before the atomic absorption spectrometry analysis. And the nitric acid was added during the process of preparation Cu(II) standard solution. In addition, the settling velocity of CPGS was measured using a glass-column, which was 40 cm in height to ensure that the terminal settling velocity could be obtained and 10.0 cm in internal diameter to minimize the wall effect on granule settling.³⁰ Afterwards, wet sludge was collected, washed several times with deionized water, and finally dried at 55 °C. Digital camera (Nikon D7000) was employed to record the appearance and sedimentation process of sludge. X-ray diffraction (XRD) patterns were collected on a TTR III diffractometer (Rigaku Co., Japan) using Cu Kα radiation (λ = 0.1542 nm). The morphology and internal structure of sludge were examined by scanning electron microscope (SEM, JSM-6360LV) and transmission electron microscopy (TEM, JEM2010).

3. Results and discussion

3.1 Effect of pH

The solution pH is known to play important roles on Cu(II) removal from wastewater, which is directly related to the form of the ions in the solution and precipitation performance. As shown in Fig. 1, the settling velocity increases by increasing pH from 7 to 11. Moreover, the settling velocity reaches the maximum 4.4 cm s⁻¹ when the pH is 11. With further increase of pH to 12, the settling velocity slightly reduces to 4.0 cm s⁻¹. OH⁻ ions is intrinsically a ligand capable of complexing with Cu²⁺ to form Cu(OH)₄²⁻.³¹ Then Cu(OH)₄²⁻ condenses and transforms to CuO, which is different from the reported pathway of Zn(II) CPGS. In zinc-bearing alkaline solution, the initial formation of ε-Zn(OH)₂, which is thermodynamically metastable with respect to ZnO.³² Fig. 1 also presents the influence of pH on Cu(II) removal efficiency, this value has been maintained at 99.3% with increasing pH. In conclusion, the optimal pH in this study is chosen as 11.

Fig. 1 Effect of pH on settling velocity and Cu(II) removal efficiency (PAM dosage of 10 mg L⁻¹; CuO seed dosage of 1.5 g L⁻¹; adding rate of 2 mL min⁻¹; stirring rate of 600 rpm).

3.2 Effect of CuO seed dosage

In the crystallization of metal oxides, seed is often desired to enhance crystallization, which has direct effect on the product morphology and properties during crystallization process. In this study, CuO, which has similar structure with expected products, was chosen as seed material. Fig. 2 shows the effect of type and dosage of CuO seed on crystallization. As noted in Fig. 2a, the settling velocity is approximately 1.4 cm s⁻¹ without CuO seed. Settling velocity increases by increasing the common CuO seed dosage from 0.5 to 2.5 g L⁻¹ and reaches the maximum value of 6.9 cm s⁻¹ when the dosage of common CuO seed is 2.5 g L⁻¹. With further increase of the common CuO seed dosage to 3.0 g L⁻¹, the settling velocity slightly reduces to 6.7 cm s⁻¹. From Fig. 2b, as dosage of nano-CuO seed increases to 2.0 g L⁻¹, the settling velocity increases and approaches a plateau of about 11.2 cm s⁻¹. Further increase of nano-CuO seed dosage, the settling velocity drops to 10.7 cm s⁻¹.

Fig. 2 Effect of (a) common CuO seed, and (b) nano-CuO seed dosage on settling velocity and Cu(II) removal efficiency (pH of 11; PAM dosage of 10 mg L⁻¹; adding rate of 2 mL min⁻¹; stirring rate of 600 rpm).

Besides controlling the number of nuclei to avoid the outbreak of nucleation, seed is used as growth nucleus, which promotes the growth of crystals.³³ Increasing the CuO seed dosage provides more surface for nucleation and size growth of crystal.^33–35 However, the growth rate of single crystal will be restrained and the mean size of products is decreased when the seed loads too much.³³ Furthermore, the settling velocity of CPGS was affected by the types of CuO seed (Fig. 2). This is because the morphology and size of seeds are important factors affected settling performance of CPGS.^36–38 Thus, there are different in the optimum of seed and settling velocity of CPGS for common CuO seed and nano-CuO seed. The common CuO was chosen as seed material because its price is significantly lower than nano-CuO's, even though the performance of nano-CuO is slightly better than common CuO's. Moreover, it can be considered to form CPGS that recycling the seed material CuO as returned sludge in the further research, which may lower the cost to a large extent. As shown in Fig. 2, changing the seed dosage has no noteworthy effect on the removal efficiency of Cu(II). Therefore, the common CuO seed was chosen as optimal seed, and the optimal dosage is 2.5 g L⁻¹.

3.3 Effect of stirring rate

Stirring will affect the mixing degree of the solution and size of the products. The collision will be enhanced with the increase of stirring intensity, then the debris particles form some new crystal nucleus making the particle size smaller. For the purpose of improving settleability of chemical precipitated sludge, the effect of the stirring rate was studied. The stirring rate was taken as 200, 400, 600 and 800 rpm in this study, and the influence of stirring rate on the settling velocity was shown in Fig. 3. As the stirring rate increases from 200 to 600 rpm, the settling velocity increases from 4.9 to 6.9 cm s⁻¹. While the stirring rate increases to 800 rpm, the settling velocity slightly reduces to 4.6 cm s⁻¹. At lower stirring rate, the local supersaturation become too high to produce a large number of small particles. The reaction becomes more complete by increasing the stirring speed. However, the destroyed probability of the crystal particles increases with further increase of the stirring rate, resulting in the decrease of grain size. Fig. 3 also presents that the stirring rate has little influence on Cu(II) removal. Consequently, 600 rpm is selected as the optimal stirring rate in this study.

Fig. 3 Effect of stirring rate on settling velocity and removal efficiency of Cu(II) (pH of 11; PAM dosage of 10 mg L⁻¹; CuO seed dosage of 2.5 g L⁻¹; adding rate of 2 mL min⁻¹).

3.4 Effect of the adding rate

High level of supersaturation is produced due to the low solubility of the resulting metal precipitate during the precipitation process. According to O. Söhnel and J. Garside,³⁹ when the level of supersaturation is very high, primary homogenous nucleation becomes the dominant mechanism by which primary particles are formed. Even if introducing seed material, it is difficult to ensure the new formation of precipitation by growth on the seed surface.⁴⁰ The reaction can be operated with a CDJP technology, which helps reduce the level of supersaturation by controlling adding rate of reactants. As shown in Fig. 4, the adding rate of reactants significantly influences on the settling velocity and Cu(II) removal from wastewater. It is found that lower adding rate is favorable to increase settling velocity. When the adding rate decreased from 8.0 to 0.2 mL min⁻¹, the settling velocity increased from 4.0 to 8.6 cm s⁻¹. When the adding rate is 8 mL min⁻¹, the level of supersaturation produced is high, leading to the production of small particles and significant technical challenges with respect to solid–liquid separation. The settling velocity of CPGS increases with adding rate decreasing. When the adding rate of 0.5 and 0.2 mL min⁻¹, the settling velocity of the CPGS is 8.1 and 8.6 cm s⁻¹, respectively. The result indicates that the level of supersaturation is closely related to flow rate. Introducing the seed and lower adding rate can reduce the supersaturation, and promote heterogeneous primary nucleation and/or secondary nucleation. The settling velocity has little difference at an adding rate of 0.2 and 0.5 mL min⁻¹, but the amount of precipitate obtained at 0.5 mL min⁻¹ is more than at 0.2 mL min⁻¹, therefore, the adding rate of 0.5 mL min⁻¹ is selected as the optimum condition.

Fig. 4 Effect of the adding rate of reactants on settling velocity and Cu(II) removal efficiency (pH of 11; PAM dosage of 10 mg L⁻¹; CuO seed dosage of 2.5 g L⁻¹; stirring rate of 600 rpm).

3.5 Effect of PAM dosage

Flocculant PAM dosage will affect the skeletal structure of the CPGS and the settling performance of flocculation. When PAM is added, particles will be aggregated through the bridging flocculation mechanism, where segments of the same polymer molecule are attached to more than one particle.⁴¹ Fig. 5 shows the effect of PAM dosage on the settling velocity and Cu(II) removal efficiency. Low settling rate (0.3 cm s⁻¹) is observed in the absence of PAM. When adding PAM, the settling velocity is gradually improved with increasing its dosage. The more polymer dose is applied, the more flocculation occurs, resulting in a higher settling velocity (Fig. 5). When PAM dosage is 20 mg L⁻¹, the settling velocity reaches the peak value of 6.5 cm s⁻¹. Thereafter, the settling velocity dropped slightly and the values is around 4.9 cm s⁻¹ because of polymer overdosed. At lower dosage, small flocs are produced, owing to less loops and tails of adsorbed polymers which were able to extend beyond the electric double layer to invoke the efficient bridging. Increasing PAM dosage, flocs with large size are formed due to the increased polymer loops and tails protruding from the particle surfaces and extending beyond the influence of the electric double repulsion. And it will improve the settling velocity because of more particles bridged and higher flocculant surface coverage.^42–44 However, high polymer dosage results in competition between the flocculants during bridging, and has a negative effect on the flocculation.⁴⁵ As seen in Fig. 5, the removal efficiency of Cu(II) does not improve with the increase of PAM dosage. As a consequence, the optimal PAM dosage is selected as 20 mg L⁻¹. This dosage of PAM ensures a sufficient free functional groups used as bridges, making more suspended solids together and building larger flocs.

Fig. 5 Effect of PAM dosage on settling velocity and Cu(II) removal efficiency (pH of 11; CuO seed dosage of 2.5 g L⁻¹; adding rate of 2 mL min⁻¹; stirring rate of 600 rpm).

3.6 Properties of chemical precipitation sludge and CPGS

Photographs of the chemical precipitation sludge sample and CPGS sample are presented in Fig. 6. Apparently, the chemical precipitation sludge (Fig. 6a) is unconsolidated, flocculent and irregular. Fig. 7 shows the sedimentation process of the chemical precipitation sludge and CPGS. It can be seen that the location of the sludge is almost the same within 1 s (Fig. 7a). Different from the chemical precipitation sludge, the CPGS (Fig. 6b) has a relatively regular shape and compact structure. More importantly, the settling velocity of these CPGS is much higher than chemical precipitation sludge (Fig. 7b). The settling velocity can reach up to 8.1 cm s⁻¹ under the optimal condition by measuring and calculating. However, the settling velocity of flocculent sludge is between the 0.17 and 0.42 cm s⁻¹.⁴⁶ Compared with the 0.42 cm s⁻¹, the settling velocity of CPGS was about 18 times higher than that of flocculent sludge. This remarkable settling performance of the CPGS will observably decrease the operation cost of sludge sedimentation process.

Fig. 6 Photographs of chemical precipitation sludge (a), CPGS (b) and their XRD patterns (filled circles CuO) (c).

Fig. 7 Sedimentation photo images of chemical precipitation sludge (a) and CPGS (b).

To further demonstrate the crystallinity of sludge samples, XRD measurement for the CuO seed and different sludge samples were conducted. The XRD patterns are given in Fig. 6c. The purity and property of CuO seed is high as evaluated by the narrow and intense peaks in the XRD pattern of CuO seed. These diffraction peaks at 2θ = 32.5° (110), 35.5° (11−1), 38.7° (111), 48.7° (20−2), 53.4° (020), 58.2° (202), 61.5° (11−3), 66.2° (31−1), 68.1° (220), 72.3° (311) and 74.9° (004), index to monoclinic CuO (space group C2/c, JCPDS 48-1548)⁴⁷ with lattice constants a = 4.688 Å, b = 3.423 Å, c = 5.132 Å. It can be observed that there are no discernible peaks in the XRD patters of chemical precipitation sludge, which is in agreement with the result of photograph (Fig. 6a) and photo image during sedimentation process (Fig. 7a). In the case of CPGS, the intensity of the CPGS diffraction peaks are observed, which are lower and wider than CuO peaks. This phenomenon possibly implies the growth of fresh precipitates on the CuO seed surfaces. Introduce seed particles into the crystallization process, fresh precipitates will tend to grow heterogeneously on the seed,⁴² which causes polycrystalline agglomerates, and furthermore lead to particles structure irregular.

3.7 CPGS formation mechanisms

To understand the mechanism of sludge formation, the products obtained at different time are studied. SEM observations (Fig. 8) and TEM images (Fig. 9) show the different morphologies of the CPGS formation in several stages and the CuO seed. As seen in Fig. 8a and 9a, the CuO seeds surfaces are smooth and have an irregular spherical shape with dense structure. From Fig. 8b, after a 15 min reaction, the surface of these CuO seeds transformed into a higher roughness, that tend to have more particles to aggregate and attach to. Meanwhile, TEM image (Fig. 9b) also shows that primary precipitates are large aggregate with dense structure. It can be noted that primary nanoparticles (grey area) pack densely into or around nucleus which was constituted of several CuO seed (black area) agglomerate together. The fresh precipitates start to grow heterogeneously on the seed surfaces due to seed-induced distortions, elastic stress accumulates in the crystallite, when the CuO seed is introduced into precipitation system.^48–51 The solution reaction of 30 min contains many tiny amorphous particles with loose structure as shown in Fig. 8c. Fig. 8d and e display the typical morphologies of the particles obtained at 45 and 60 min during the CPGS formation process, respectively. It demonstrates that precipitate morphology, structure and size are not materially different. Only more precipitates appeared (Fig. 8c–e) compared to the beginning stage (Fig. 8a and b). In addition, TEM images (Fig. 9c–e) show the slightly increase of particles size, and the sludge morphology transform from needle to lamellar.

Fig. 8 SEM images of the CuO seed and sludge samples of CPGS in the presence of a seed proceeds in several stages: (a) CuO seed; (b) 15 min; (c) 30 min; (d) 45 min; (e) 60 min; (f) rapid stirring 30 s; (g) slow stirring 15 min; (h) quiescent aging 30 min.

Fig. 9 TEM images of the CuO seed and sludge samples of CPGS in the presence of a seed proceeds in several stages: (a) CuO seed; (b) 15 min; (c) 30 min; (d) 45 min; (e) 60 min; (f) rapid stirring 30 s; (g) slow stirring 15 min; (h) quiescent aging 30 min.

When a certain dosage of PAM is added, tiny particles are almost disappeared, and most of the particles have larger size (Fig. 8f and 9f) than those in precipitation process (Fig. 8b–e and 9b–e). However, the internal structure of sludge is still loose and fragile. When PAM added, flocculation of particles may occur by polymer bridging, charge compensation or neutralization, due to PAM floc structures having an open structure with large voids.⁴¹ Followed by a slow stirring (100 rpm) for 15 min, the flocs are less robust and easily destroyed as shown in Fig. 8g and 9g.

Fig. 8h displays the morphologies of the sludge obtained by quiescent aging 30 min, which demonstrates that smaller aggregates with dense structure obtained compared to the beginning of introducing PAM. Moreover, the result in Fig. 9e confirms that sludge has a regular shape, high degree of sphericity with smooth surface, consistent with the characteristics of granular sludge.⁵² In summary, the addition of PAM is necessary for forming CPGS by bridging flocculation arising from its high molecular weight and linear chain structure.

Based on the above results, the formation mechanism of the CPGS during precipitation process can be composed of three steps (Fig. 10): (1) formation of fresh precipitates on the CuO seed surfaces using the CDJP at pH of 11.0; (2) flocculation of particles that a proper amount of PAM flocculant is added to obtain flocs, the CPGS preliminary formation; (3) shear provides the loading force essential for rearrangement and re-conformation of the flocculant molecule, which leads to more effective bridging between particles. The key to formation of the CPGS is adding seeds and application of the CDJP. Using CDJP will increase the size of the primary precipitates and prevent the homogeneous nucleation by controlling the solution supersaturation.⁵³ Furthermore, the CuO seed is an indispensable part of growth nuclei during successful formation of the CPGS.

Fig. 10 Formation mechanism of CPGS.

4. Conclusions

In this study, the process of copper wastewater treatment by forming CPGS was investigated to improve sludge settling performance. The results show that the optimal pH, CuO seed type and dosage, stirring rate, adding rate and PAM dosage is 11, common CuO, 2.5 g L⁻¹, 600 rpm, 0.5 mL min⁻¹ and 20 mg L⁻¹, respectively. Furthermore, forming CPGS is easy to realize by introducing the CuO seed and using CDJP technology in precipitation process. The CuO seed play an important role of nuclei growth. Reducing the level of supersaturation can prevent primary homogenous nucleation through controlling adding rate using CDJP technology. More importantly, the settleability of CPGS can be improved by 18 times as against that of flocculent chemical precipitation sludge. Meanwhile, the strategy of CPGS is essentially changing the flocculent nature of chemical precipitation sludge.

Acknowledgements

This work was supported by National Natural Science Foundation of China (51504299); Science and Technology Program for Public Wellbeing (2012GS430201); The Key Project of Science and Technology of Hunan Province, China (2012FJ1010) for financial support.

Notes and references

1. M. R. Awual, Chem. Eng. J., 2015, 266, 368–375.
2. Y. Ding, S. Z. Shen, H. Sun, K. Sun and F. Liu, Sens. Actuators, B, 2014, 203, 35–43.
3. M. R. Awual, T. Yaita, S. A. El-Safty, H. Shiwaku, S. Suzuki and Y. Okamoto, Chem. Eng. J., 2013, 221, 322–330.
4. J. Gao, F. Liu, P. Ling, J. Lei, L. Li, C. Li and A. Li, Chem. Eng. J., 2013, 222, 240–247.
5. J. M. Sun, C. Shang and J. C. Huang, Environ. Sci. Technol., 2003, 37, 4281–4287.
6. A. Jawor and E. M. V. Hoek, Environ. Sci. Technol., 2010, 44, 2570–2576.
7. I. Dobrevsky, M. Dimova-Todorova and T. Panayotova, Desalination, 1997, 108, 277–280.
8. M. A. Hossain, H. H. Ngo, W. S. Guo and T. V. Nguyen, Bioresour. Technol., 2012, 113, 97–101.
9. D. L. Boccelli, M. J. Small and D. A. Dzombak, Environ. Sci. Technol., 2005, 39, 6501–6507.
10. F. Gode and E. Pehlivan, J. Hazard. Mater., 2006, 136, 330–337.
11. C. W. Cheung, C. K. Chan, J. F. Porter and G. McKay, Environ. Sci. Technol., 2001, 35, 1511–1522.
12. D. C. K. Ko, J. F. Porter and G. McKay, Environ. Sci. Technol., 2001, 35, 2797–2803.
13. A. C. Scheinost, S. Abend, K. I. Pandya and D. L. Sparks, Environ. Sci. Technol., 2001, 35, 1090–1096.
14. M. Peld, K. Tonsuaadu and V. Bender, Environ. Sci. Technol., 2004, 38, 5626–5631.
15. J. L. Huisman, G. Schouten and C. Schultz, Hydrometallurgy, 2006, 83, 106–113.
16. F. Fu and Q. Wang, J. Environ. Manage., 2011, 92, 407–418.
17. A. L. Mackie and M. E. Walsh, Water Res., 2012, 46, 327–334.
18. M. Citeau, J. Olivier, A. Mahmoud, J. Vaxelaire, O. Larue and E. Vorobiev, Water Res., 2012, 46, 4405–4416.
19. R. D. Ambashta and M. Sillanpaa, J. Hazard. Mater., 2010, 180, 38–49.
20. Y. J. Tu, C. K. Chang, C. F. You and S. L. Wang, J. Hazard. Mater., 2012, 209–210, 379–384.
21. W. R. Knocke and R. T. Kelley, J.–Water Pollut. Control Fed., 1987, 59, 86–91.
22. W. A. Parsons and H. Heukelekian, Ind. Eng. Chem., 1954, 46, 1503–1507.
23. X. Feng, J. Deng, H. Lei, T. Bai, Q. Fan and Z. Li, Bioresour. Technol., 2009, 100, 1074–1081.
24. M. Saha, C. Eskicioglu and J. Marin, Bioresour. Technol., 2011, 102, 7815–7826.
25. J. Beltran Heredia and J. Sanchez Martin, J. Hazard. Mater., 2009, 165, 1215–1218.
26. F. M. Pang, P. Kumar, T. T. Teng, A. K. Mohd Omar and K. L. Wasewar, J. Taiwan Inst. Chem. Eng., 2011, 42, 809–815.
27. X. Yan, Q. Li, L. Chai, B. Yang and Q. Wang, Chemosphere, 2014, 113, 36–41.
28. W. Li, P. Zheng, L. Wang, M. Zhang, H. Lu, Y. Xing, J. Zhang, R. Wang, J. Song and A. Ghulam, Bioresour. Technol., 2013, 142, 683–687.
29. X. Yan, L. Chai and Q. Li, Sep. Sci. Technol., 2013, 48, 1442–1449.
30. H. F. Lu, P. Zheng, Q. X. Ji, H. T. Zhang, J. Y. Ji, L. Wang, S. Ding, T. T. Chen, J. Q. Zhang, C. J. Tang and J. W. Chen, Bioresour. Technol., 2012, 123, 312–317.
31. F. Meng and S. Jin, Nano Lett., 2012, 12, 234–239.
32. X. Yan, L. Chai, Q. Li, L. Ye, B. Yang and Q. Wang, CrystEngComm, 2016, 18, 924–929.
33. D. C. Huang, W. Liu, S. K. Zhao, Y. Q. Shi, Z. X. Wang and Y. M. Sun, Chem. Eng. J., 2010, 156, 360–365.
34. S. H. Lin and M. C. Lo, J. Hazard. Mater., 1998, 63, 211–222.
35. N. Doki, N. Kubota, A. Sato and M. Yokota, AIChE J., 1999, 45, 2527–2533.
36. Y. Deng, J. Zhao, Q. Li, X. Xu, H. Lin and Y. Li, CrystEngComm, 2014, 16, 5184.
37. T. Kawano and H. Imai, Colloids Surf., A, 2008, 319, 130–135.
38. M. Huang, Y. Yan, W. Feng, S. Weng, Z. Zheng, X. Fu and P. Liu, Cryst. Growth Des., 2014, 14, 2179–2186.
39. O. Söhnel and J. Garside, Precipitation: basic principles and industrial applications, Butterworth-Heinemann, Oxford, 1992.
40. T. P. Mokone, R. P. van Hille and A. E. Lewis, Water Res., 2012, 46, 2088–2100.
41. Y. D. Yan, S. M. Glover, G. J. Jameson and S. Biggs, Int. J. Miner. Process., 2004, 73, 161–175.
42. A. R. Heath, P. A. Bahri, P. D. Fawell and J. B. Farrow, AIChE J., 2006, 52, 1987–1994.
43. P. Mpofu, J. Addai-Mensah and J. Ralston, Int. J. Miner. Process., 2003, 71, 247–268.
44. Y. Zhou and G. V. Franks, Langmuir, 2006, 22, 6775–6786.
45. Y. L. Luo, Z. H. Yang, Z. Y. Xu, L. J. Zhou, G. M. Zeng, J. Huang, Y. Xiao and L. K. Wang, J. Hazard. Mater., 2011, 189, 69–75.
46. Y. Wu, R. Wang and Y. Peng, Granular Sludge Technology of Wastewater Treatment: Principles and Applications, China Architecture & Building Press, Beijing, 2011.
47. J. Y. Xiang, J. P. Tu, L. Zhang, Y. Zhou, X. L. Wang and S. J. Shi, J. Power Sources, 2010, 195, 313–319.
48. U. Dusek, G. P. Frank, L. Hildebrandt, J. Curtius, J. Schneider, S. Walter, D. Chand, F. Drewnick, S. Hings, D. Jung, S. Borrmann and M. O. Andreae, Science, 2006, 312, 1375–1378.
49. A. Cacciuto, S. Auer and D. Frenkel, Nature, 2004, 428, 404–406.
50. E. Allahyarov, K. Sandomirski, S. U. Egelhaaf and H. Loewen, Nat. Commun., 2015, 6, 1–9.
51. S. Sathyamoorthy, G. D. Moggridge and M. J. Hounslow, Cryst. Growth Des., 2001, 1, 123–129.
52. J. E. Schmidt and B. K. Ahring, Biotechnol. Bioeng., 1996, 49, 229–246.
53. M. Jitianu and D. V. Goia, J. Colloid Interface Sci., 2007, 309, 78–85.

---