Remarkable phosphate removal from saline solution by using a novel trimetallic oxide nanocomposite

Phosphate removal is an important measure to control eutrophication in aquatic environments, as it inhibits algal bloom. Salinity exists in these media along with high phosphate and currently available phosphate removal methods function poorly under this condition. In this study, the main objective is to fabricate a nanocomposite to improve and accelerate phosphate removal from saline solutions. To achieve this goal, Fe3O4/ZnO and a novel nanoadsorbent, Fe3O4/ZnO/CuO, were synthesized. Their characteristics were determined using FE-SEM, EDX, FT-IR, and XRD analyses, and their capability to adsorb phosphate from saline solutions was investigated and compared. The overall results suggest that the trimetallic oxide nanocomposite has great potential for the efficient removal of phosphate, in comparison with Fe3O4/ZnO. Experiments showed that Fe3O4/ZnO/CuO exhibited a remarkable sorption capacity of 156.35 mg P/g, fast sorption kinetic, strong selectivity for phosphate even in the presence of a high concentration of salinity (60 mg/L), and a wide applicable pH range of 3–6. Furthermore, using Fe3O4/ZnO/CuO, even a low dosage of 0.1 g/L was sufficient to reach an adsorption efficiency of 96.13% within 15 min compared to Fe3O4/ZnO (80.47% within 30 min). Moreover, the pseudo-second-order kinetic model best described the experimental adsorption data for both nanocomposites.

High amounts of salt are used in oil, textile, and leather industries. During their processes, a large amount of saline wastewater is produced, which finally enters surface waters. Moreover, high amounts of NaCl and Na 2 SO 4 salts exist naturally in free waters as well (Stewart ). According to the results of various studies, the phosphorous removal efficiency decreases significantly with an increase in the amount of salinity in the solution (Lefebvre & Moletta ). Therefore, it is of high significance to find a method that functions well under this condition.
Thus far, conventional approaches such as chemical precipitation and biological treatment are incapable of removing low concentrations of phosphate; therefore, other methods are required so as to overcome the mentioned problems (Wu et al. ). Due to the simplicity in design, the effectiveness even at low concentrations of phosphorus, and the recycling capability, the adsorption method has been widely accepted as a preferable approach for phosphate removal (Wu et  () achieved 44.14 mg P/g adsorption capacity by using Ti-NS nanocomposite at pH 6.5-7.5, nanocomposite dosage of 500 mg/L, initial phosphate concentration of 10 mg/L, and at 25 C. In Zhou et al.'s () study, the adsorption capacity of 37.86 mg P/g was reported by the use of nHFZO@I402 nanocomposite at pH 7, nanocomposite dosage of 750 mg/L, phosphate concentration of 5 mg/L, and at 25 C. In another study, Wu et al. () achieved the adsorption capacity of 83.5 mg P/g by using La(OH) 3 /Fe 3 O 4 nanocomposite at pH 4.3-6, nanocomposite dosage of 100 mg/L, initial phosphate concentration of 0.5-15 mg/L, and at 23 C. In Nodeh et al.'s () research, the adsorption capacity of 116.28 mg P/g was achieved by using MG@La nanocomposite at pH 6, nanocomposite dosage of 100 mg/L, initial phosphate concentration of 50 mg/L, and at 25 C.
Moreover, the recyclability of an adsorbent is of great importance to evaluate its performance for practical applications (Hu et al. ). Nanoparticles are difficult to separate after usage in wastewater treatment, and commonly used adsorbent recycling systems suffer from several disadvantages. While separation methods such as centrifuge consume a lot of energy, and filtration is susceptible to clogging, magnetic separation is a faster and more effective method (Wu et al. ; Cai et al. ). Accordingly, the combination of magnetic nanoparticles (such as Fe 3 O 4 ) with other nanomaterials makes the separation and recycling of the synthesized adsorbent more facile.
Molybdovanadate reagent was also used for analyzing the concentration of phosphate in the solution, all of which were purchased from Merck.

Equipment
The equipment used in this research was Hach DR4000U

Synthesis of nanocomposites
To synthesize the nanoadsorbents, the following steps were taken for each of the nanocomposites.

Preparation of Fe 3 O 4 /ZnO/CuO
CuO nanoparticles were synthesized using the hydrolysis method (Tran & Nguyen ). To prepare these nanoparticles, 0.6 g of copper acetate (II) was dissolved in 50 ml of distilled water on a stirrer at 100 C, and 0.01 g of NaOH was added. Then, 0.1 g of the synthesized Fe 3 O 4 /ZnO nanocomposite was dispersed in 25 ml of distilled water, added to the copper solution, and stirred for 1 h at 80 C. After hydrolysis, the precipitate in the solution was washed several times with distilled water and ethanol, and then the resulting substance was dried at 200 C for 2 h.

Methods
Phosphate adsorption from the synthesized wastewater using  Table 1.
A 100 ml glass beaker was used for phosphate removal examinations at a temperature of 25 ± 0.5 C, and all experiments were performed three times, the error rate of which was less than 1.5%. After each test, the used nanocomposite was separated with a magnet to avoid a possible metal leakage. The remaining liquid in the beaker was used to determine the concentration of residual phosphate in the solution by the colorimetric method using the UV-Vis spectrophotometer and the molybdovanadate reagents (0.5 ml) (Eaton et al. ). Then, the phosphate adsorption efficiency was calculated using the following equation: where C 0 (mg/L) is the initial concentration of the phosphate solution, C e (mg/L) is the concentration of phosphate solution after adsorption, and E (%) is the efficiency of phosphate adsorption.
where C 0 and C e (mg/L) are initial and equilibrium concentrations of the phosphate solution, respectively, V (L) is the volume of the phosphate solution, and M s (g) is the mass of the utilized nanocomposite.
After obtaining the optimum conditions for phosphate removal, phosphate adsorption tests were conducted in the presence of NaCl and Na 2 SO 4 salts, which are the predominant salts in both industrial wastewater and free waters (Stewart ). Then, the amount of adsorption efficiency was determined.
Adsorption kinetics were also investigated by pseudofirst-order and pseudo-second-order kinetic equations (Equations (3) and (4)) (Ho & McKay ): where K 1 (L/h) is the rate constant of the pseudo-first-order model, K 2 (g/mg h) is the rate constant of the pseudosecond-order model, and q and q e (mg/g) are the amount of phosphate adsorbed to the adsorbent at time t (min).
To determine pH at the point of zero charge (pH pzc ) for each nanocomposite, the following approach was used.
First, 0.01 M NaCl solution was added to sealed vials, and the initial pH of each vial was adjusted to the values between 2 and 10 using pH regulator solution. 30 mg of each nanocomposite was added to vials with different initial pH and shaken for 48 h at room temperature. Using a pH meter, the final pH of each solution was determined. The initial and final graphs were plotted, and pH pzc is the

Magnetic adsorbent characteristics
To determine the characteristics of the synthesized nanocomposites, FE-SEM, EDX, FT-IR, and XRD analyses were performed, the results of which are presented in Figures 1-4.

FE-SEM analysis
To

XRD analysis
XRD patterns were generated using Cu Kα radiation (40 kV, 40 mA) to identify the crystalline components of both adsorbents. All XRD patterns were obtained over a 2ϴ range of 5-80 . Figure 4(a) and 4(b) illustrates XRD patterns of where pK 1 ¼ 2.12, pK 2 ¼ 7.21, and pK 3 ¼ 12.67, respectively respectively. At lower pH (under pH pzc ), due to protonation, both nanocomposite surfaces were positively charged, and the dominant phosphate ions H 2 PO À 4 were more easily adsorbed on the adsorbents' surface. Therefore, the adsorption efficiency was higher in acidic pH. As pH increases (more than pH pzc ), through deprotonation, the positively charged surface of both of the nanocomposites gradually changes into negative. As a result, the adsorption efficiency

The effect of the initial concentration of nanocomposite
To investigate the effect of the adsorbents dosage, different amounts of them were tested, the results of which are presented in Figure 6. As can be observed in Figure 6

Effect of initial phosphate concentration
To determine the optimum phosphate concentration, different amounts of this contaminant were examined, the results of which are presented in Figure 7. Based on Besides, by comparing the graphs in Figure 7 i.e., PO 3 -  95, 66.12, 56.31, 57.4, 48.6, and 39.81%, respectively, while this efficiency for trimetallic oxide nanocomposite was 90.74, 89.32, 88.02, 85.62, 83.45, and 80.37%, respectively. The phosphate removal efficiency in the presence of the mentioned concentrations of Na 2 SO 4 salinity while using Fe 3 O 4 /ZnO was 76. 4,75.3,55.3,53.9,49.5,and 43.72%,respectively,and this efficiency using Fe 3 O 4 /ZnO/CuO was 90.13,88.48,87.23,86.33,83.69,and 81.05%, respectively. The decrease in the contaminant removal efficiency in the presence of salinity can be due to the competition that occurs between ions. Similar results have been reported in several studies. In Hong et al.'s () study, Fe 3 O 4 @SiO 2 nanocomposite was used in the presence of HCO À 3 anion, which reduced the phosphate removal efficiency by 29%. In a study by Hu et al. (), Zr@MCS nanoadsorbent was used for phosphate removal in the presence of SO 2À 4 anion, which reduced the adsorption efficiency by more than 15%.

Equilibrium and kinetic adsorption studies
To evaluate the rate of phosphate adsorption on the surface of the nanocomposites, pseudo-first-order and pseudosecond-order kinetic models were applied to the adsorption data. Their results under two situationsthe presence and absence of salinityare presented in Table 3. The pseudosecond-order kinetics has a higher correlation coefficient (R 2 ) than the pseudo-first-order (closer to 1) in the absence of salinity in the system. So, it can be concluded that phosphate adsorption kinetics follows the pseudo-second-order Adsorption kinetics was also applied on phosphate removal in the presence of salinity (NaCl and Na 2 SO 4 salts with a concentration of 60 mg/L) in the system. The concentration of 60 mg/L was chosen due to its substantial effect on the reduction of phosphate removal efficiency, the results of which are presented in

REGENERATION AND REUSABILITY OF THE BIMETALLIC AND TRIMETALLIC OXIDE NANOCOMPOSITES
To investigate the reusability of both nanocomposites, the sorption-desorption experiments were repeated for 10 cycles. The bimetallic and trimetallic oxide nanocomposites were dried at room temperature and then regenerated by 1 mol/L NaOH solution with a contact time of 20 min. This desorption reagent could efficiently regenerate both of the adsorbents due to the tendency to replace OH À particles with H 2 PO À 4 particles. The NaOH-regenerated nanocomposites were reused for phosphate removal under optimum conditions. As illustrated in Figure 9, the results showed that with the use of Fe 3 O 4 /ZnO, the phosphate removal efficiency had a 2-3% decrease in each cycle and after the 10th cycle, the efficiency declined up to 20%. On the other hand, by using the modified trimetallic oxide nanocomposite, the phosphate adsorption efficiency had less than 1% decrease after each cycle, while, in the 10th cycle, it declined to less than 10%.

CONCLUSION
In this study, two magnetically separable Fe 3 O 4 /ZnO and modified Fe 3 O 4 /ZnO/CuO nanocomposites were synthesized and characterized. Then, their ability to adsorb phosphate and the effect of salinity on their performance were investigated and compared. According to the results, the trimetallic oxide nanocomposite had a high capacity to adsorb phosphate in the absence and particularly in the presence of salinity, while Fe 3 O 4 /ZnO showed less resistance to salinity and had a lower phosphate removal efficiency.
Results demonstrated about 50% increase in phosphate adsorption rate by using Fe 3 O 4 /ZnO/CuO and its significant improvement in comparison to Fe 3 O 4 /ZnO, due to the decrease in reaction time and energy, and the increase in the phosphate removal efficiency. Regeneration and reusability studies showed that the trimetallic oxide nanocomposite had a notable improvement in comparison to Fe 3 O 4 /ZnO due to its negligible adsorption efficiency reduction after

DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.