Synthesis of hydrated ferric oxide on cation exchange resin for phosphate and hardness removal in water

In this study, a potential adsorbent was synthesized from iron salt and cation exchange resin (FeOOH@CR) and applied for phosphate adsorption in batch experiments. The characteristics of FeOOH@CR materials before and after phosphate adsorption were determined by FTIR, XRD, and SEM. The factors affecting the adsorption process such as reaction time, solution pH, material dosage, concentration, temperature, and competing ions were tested. Kinetic, thermodynamic and isothermal models of the adsorption process were applied to study the nature of the adsorption process. The properties of phosphate adsorption, effect of competitive ions and material reusability were also examined. Results showed that the adsorption time reached equilibrium after 48 h and the suitable adsorption condition was found at solution pH of 6.5, material dosage of 5 g/L. In addition, the durability of the material after 5 times of regeneration was investigated with the remained adsorption ability of about 55% as compared to the original one.


Introduction
The anthropogenic nutrients in the aquatic environment should be reduced to protect the water sources for human uses and reduce eutrophication (e.g., algae growth) in the water environment, which decreases the oxygen content in the water and damages aquatic life conditions [1,2]. The concentration of phosphorus have been significantly increased in water due to excess fertilizer use and also from the discharge of municipal wastewater. Therefore, controlling phosphorus is necessary for the country's ecological survival [3,4]. Phosphate concentration limit in domestic wastewater in Vietnam is less than or equal to 6 mgP-PO4 3-/L (Column A, QCVN 14:2008/BTNMT). Nevertheless, to meet increasingly stringent requirements, methods such as chemical precipitation, biologic processes, and adsorption by functionalized materials are proposed as an alternate option that is currently in commercial usage around the world [26]. The fact demonstrates that ion exchange resins are frequently used in the treatment of water hardness, lowering costs and improving treatment efficiency.
Recent research have revealed that hardness and phosphate treatment can be done separately, making treatment systems more complex, with a larger footprint and higher costs. In this study, iron oxyhydroxide on cation exchange resins was used to create a new FeOOH@CR material for simultaneous removal of phosphate and cations such as Ca 2+ and Mg 2+ in this study. This research demonstrates a vision for the material's practical use in the treatment of phosphate at low concentrations and water with significant hardness.

Chemicals
Lab-grade chemicals used in the study such as FeCl3.6H2O, NH4Cl, NH3 (28-34 vol.%), and NaCl were from China, and HCl, KH2PO4, KCl, K2SO4, MgCl2, KNO3, and CaCl2 were from Merck. Cation exchange resin H + (225H) was from India and deionized water was taken from the laboratory. The solutions contain phosphate (the concentration of mgPO4 3in the solution of the experiments is calculated by the concentration of mgP in PO4 3in the solution), calcium, magnesium were prepared by dissolving KH2PO4, CaCl2, MgCl2 salts into deionized water.

Material synthesis and characterizations
FeOOH@CR material was prepared according to a process in the previous publications [5,22,23]. The cation-exchange resin hybrid adsorbent was dispersed with ferric oxyhydroxide particles prepared by the following procedure. At first, 2 mL of HCl solution was added into 0.2 L of 12.5% w/w FeCl3.6H2O solution and the mixture was stirred for 5 min with a stirrer. After the mixture was completely dissolved, the solution had a low pH of 3. Next, 10 g of dry H + ion exchange resins (225H of India) was added to the solution and the solution was stirred at 500 rpm for 20 min. The solids were then separated from the mixture and washed with deionized water 3 times, followed by agitating for 10 min at 500 rpm in 200 mL of deionized water containing 1 g of NH4Cl and 2 mL of NH3 solution (28-34 vol. %). Subsequently, the material was separated from the solution and washed with deionized water 3 times. Finally, the material was dried in the open air until it turns red.
Scanning electron microscopy -SEM (JCM-7000, JEOL Ltd, Japan) was used to examine the morphology of the material. Energy dispersive X-ray analysis EDX (JCM-7000, JEOL Ltd, Japan) was used to determine various elemental compositions. Fourier-transform infrared spectroscopy -FTIR (Alpha, Bruker, Germany) was used to determine the surface chemistry. X-ray diffraction -XRD (D2 Phaser Benchtop X-ray Powder Diffraction, AERIS, Malvern Panalytical Ltd, Netherlands) was used to evaluate the crystal phase and crystalline structure of the materials .

Adsorption tests
Batch adsorption studies were used to investigate the adsorption of phosphate and cations (i.e. Mg and Ca). In the first test, 0.25 g of material was added to 50 mL of phosphate solution (20 mgPO4 3-/L) and allowed to adsorb for up to 80 h. The material was then removed from the solution, and a sample of the solution was collected for UV-Visible spectroscopy analysis of the phosphate concentration (V-730 UV, JASCO International Co.Ltd, Japan). The investigated influencing parameters include: contact time (0 -80 h), adsorbent dosage (2 -30 g/L), solution pH (2 -12), and competing ions in solution (sulfate, bicarbonate, chloride, and nitrate), initial phosphate concentration (10 -50 mg/L) and temperature (5 -40 o C). The phosphate adsorption capacity (in terms of phosphorous, (mgPO4 3-/g or mgP/g ) was calculated as the following equation [5,22] Where Co and Ce are initial and equilibrium concentrations of phosphate in terms of phosphorous (mgPO4 3-/L). V is the volume of solution (L) and m was the mass of the material (g). The simultaneous adsorption of phosphate and cations (Mg and Ca) was achieved by adding 0.25 g of resin into 50 mL of a solution containing 20, 100, and 100 mg/L of phosphate, calcium, and magnesium, respectively. The other steps of the experiment were similar to the previous phosphate adsorption experiment. The solution was then taken for measurement of phosphate content using UV-Visible spectroscopy and calcium and magnesium content using atomic absorption spectroscopy (AAS, Perkin Elmer Aanalyst 400, PerkinElmer, Inc., USA).

Characterizations of materials
The morphology and surface elemental composition of the materials were examined by SEM and EDXmapping, respectively ( Figure S1, S2, and S3 of Supplementary Data). Accordingly, the composition of the elements on the surface of the materials is summarized in Table 1 with the main components of C, O, Fe, Ca, Mg, and Al. Phosphate was only detected in the structure of the FeOOH@CR material after phosphate adsorption ( Figure S2), but not in those of the resin ( Figure S1) and fresh FeOOH@CR material ( Figure  S3). This result showed that the process was successful when iron was present on the surface of the material. Besides, phosphate was also adsorbed on the surface of the material after the experimental process.  41]. In this study, the surface chemical structure of FeOOH@CR before and after adsorption was examined to clarify the adsorption mechanism occurring on the surface of the material. The FTIR spectra of FeOOH (i.e., goethite and lepidocrocite) were also provided for reference purposes. As can be observed in Figure 1, all samples have a peak at wavenumber of 3390 cm -1 , which is assigned to the H-O-H vibration of the hydrated group. The peaks at the region of 789 and 880 cm -1 characterize for Fe-OH-Fe vibration [41][42][43] while the peaks at 466 and 746 cm -1 are attributed to the Fe-O-H vibrations [42,43]. The absorption bands at 3133 and 3384 cm -1 are related to the stretching vibrations of the OH groups. The peak of phosphate ranges from 1000-1100 cm −1 [44][45][46]. In this study, the dose of phosphate was much smaller than iron and the peak of phosphate was also located at the same position as the peak of iron (lepidocrocite and goethite reference [42,43]). These results are similar to those reported in the previous studies [42,43].  Figure 1. FTIR spectra of (1) FeOOH@CR after phosphate adsorption, (2) fresh FeOOH@CR, and (3) lepidocrocite and goethite reference [42,43].

Adsorption of phosphate onto FeOOH@CR
During the 80-h experiment, Figure 3 demonstrates that the adsorption capacity of FeOOH@CR increased with increasing adsorption time. With an adsorption capacity of 3.10 mgPO4 3-/g, it reached equilibrium after about 48 h, which is similar to that described in the literature. Using FeOOH@CR material, there are three primary stages in phosphate adsorption, including the first stage of quick adsorption in the first 24 h, the second stage of slow adsorption in the next 24 h, and the last stage of equilibrium after 48 h of adsorption.  Table 2, the kinetics of phosphate adsorption onto FeOOH@CR were investigated using three kinetic models of pseudo-first-order, pseudo-second-order, and intra-particle diffusion models. All three models can be utilized to describe phosphate adsorption based on the correlation coefficient (R 2 > 0.9). However, because it has the greatest R 2 value of 0.9848 ( Figure S4) with similar predicted and tested adsorption capacities, the pseudo-first-order model is the best suitable adsorption model for phosphate adsorption utilizing FeOOH@CR. At a phosphate concentration of 20 mg/L, the K adsorption rate constant was calculated to be 0.0641 (g.mg -1 .h -1 ).   The adsorption capacity of materials is affected by solution pH, which is an important parameter. Figure  4 shows the phosphate adsorption capability of the materials from pH 2 to 12. It was also shown that the adsorption capacity increased steadily from pH 2 to pH 6.5, peaking at 2.78 mgPO4 3-/g. In addition, the point of zero charge (pHpzc) of the material was determined to be 6.5 ( Figure S5). When the pH is less than 6.5, the material surface has a positive charge, which increases the adsorption of negatively charged phosphates in the forms of HPO4 2and H2PO4 - (Reactions 1, 2, and 3), and are retained by iron oxyhydroxide by forming a complex inside the sphere. Because HPO4 2forms stronger amphoteric complexes than H2PO4 -(Reactions 6 and 7), reducing the pH causes more phosphate to stay in the form of H2PO4 -, which has a lower negative charge, making the adsorption process more difficult. Furthermore, the solution with low pH reduces the electrostatic attraction between the positively charged iron oxyhydroxide groups and the phosphate ions by reducing the negative charge of phosphate from divalent to monovalent. When the pH was higher than 6.5, the adsorption capacity gradually dropped until it was just 2.12 mgPO4 3-/g at pH 12. FeOOH is deionized and negatively charged at these high pH levels (Reaction 4 and 5). As a result, phosphate adsorption is inhibited by electrostatic repulsion and Donnan co-ion exclusion [48,49]. The adsorption capacity of phosphate is influenced by the dose of FeOOH@CR adsorbent, as shown in Figure 5. The adsorption capacity declines as the adsorbent dose increases, whereas the adsorption efficiency and the output water quality gradually improves. The phosphate adsorption must meet the QCVN 14:2008/BTNMT (Column A) standard with an output phosphate content of less than 6 mgP/L. When the adsorbent dose was raised in the range of 2 -10 g/L, the phosphate adsorption capacity declines drastically, then slowly in the range of 10 -30 g/L. For the following studies, the optimum adsorbent dosage was chosen at 5 g/L.  The link between the phosphate on the materials surface and its equilibrium concentration in the solution at constant temperature was described using the Langmuir and Freundlich isotherm models [50,51]. Table  3 shows the adsorption isotherm parameters. Although both the Langmuir and Freundlich models are suitable for describing phosphate adsorption at 30 and 40 ° C, the Langmuir model has a better correlation coefficient with R 2 of 0.9974 at 30 o C, making it more favorable. The KL value, which is the Langmuir constant that represents the adsorbent's affinity for its surface [52][53][54], was estimated from the equation to be 0.26552 and 0.3705 L/mg at 30 °C and 40 o C, respectively. The thermodynamic process of adsorption, on the other hand, is stated using equilibrium constants at different temperatures (such as 278, 293, 303, and 313 0 K) utilizing factors such as Gibbs free energy (UG), entropy (US), and enthalpy (UH). The Gibbs free energy (UG) is a measure of whether a substance is physically or chemically adsorbed. The UG values between -20 and 0 kJ/mol indicate a physical adsorption process while those between -80 and -400 kJ/mol indicate chemisorption [55]. At 278, 293, 303, and 313 o K, the UG was estimated to be -0.845, -4.389, -6.751, and -9.114 kJ/mol, showing that the adsorption process was spontaneous at all investigated temperatures [53]. Van der Waals forces, hydrogen bonds, ionic pairs, and other polar and nonpolar interactions should all play a part in the interaction between phosphate ions and the FeOOH@CR surface during the adsorption process [53,55,56]. The positive value of UH indicates that this is an endothermic adsorption process with 64.84 (kJ/mol). The value of US is higher than 0, indicating that the affinity and contact between the adsorbate and the adsorbent of the adsorption process are increasing [56]. This can also be explained that ΔS and ΔH were the key factors for determining the overall sign and magnitude of the binding free energy (ΔG). Only when the free energy change was negative, the adsorption can occur spontaneously. And the positive ΔS along with negative ΔH contributed favorably to the overall binding free energy (ΔG) [53]. Besides, the positive value of ΔS suggests increased randomness at the solid/solution interface during the adsorption [55,57,58].  Figure 7 shows the impacts of anions on the phosphate adsorption capacity of FeOOH@CR material. Anions such as Cl -, SO4 2-, NO3 -, and HCO3commonly coexist with PO4 3in water and wastewater. The anions HCO3 -, at a concentration of 100 mg/L, had the greatest impact on the phosphate adsorption capacity. Other anions, such as Cl -, NO3 -, and SO4 2had a lower effect than HCO3on the adsorption process, demonstrating the specific interaction and adsorption of FeOOH@CR material with HCO3and PO4 3anions. As seen in Figure 8, the presence of cations such as calcium and magnesium in the solution increased the phosphate adsorption. The fundamental reason for this is that the presence of Ca and Mg increases the positive charge of the material surface, which makes negative phosphate anions easier to adsorb. The material was also capable of eliminating Ca 2+ from 210 to 128 mg/L and Mg 2+ from 202 to 142 mg/L at the same time. One probable explanation is that after treating the resin with NH3 solution, iron and other cations dissociate from their ionic bonds with the resin, forming oxides/hydroxides particles inside the pores. As a result, the ion exchange resin is still capable of accepting new Ca 2+ and Mg 2+ cations as they arrive in the solution, positively charging the surface to aid phosphate adsorption. In the addition of Ca 2+ and Mg 2+ , the phosphate adsorption capacity increased by 1.21 times, reaching 3.652 mgPO4 3-/g in Table 5.  Besides, this material showed that the total adsorption capacity of phosphate (3.625 mg/g) and hardness (28.4mg/g) was 32.025 mg/g compared to a commercial product (INDION MB-6 SR, Ion Exchange (India) Ltd., India) which had a total adsorption capacity of phosphate (4.4 mg/g) and hardness (8.6 mg/L) was 13 mg/g. This can be seen that the adsorption capacity of the material in this study was 2.46 times higher than that of the INDION MB-6 SR material. INDION MB-6 SR was a commercially available material on the market, it had the ability to remove both cations and anions in water.

Application to real wastewater, durability, and comparisons with other studies
After that, the FeOOH@CR material was used to treat real wastewater. As shown in Figure S6 Figure 9 shows five cycles of adsorption-regeneration that were used to verify the material durability. After five reuse cycles, the adsorption capacity was only 1.75 mgPO4 3-/g, which was approximately 56.68% of the capacity of the fresh material. Table 6 compares the adsorption capacity of various nanomaterials, showing that FeOOH@CR is potential for phosphate application.

Conclusions
The FeOOH@CR material was successfully synthesized and applied for phosphate and hardness removal in water. The equilibrium adsorption time was determined to be 48 h while the suitable pH was around 6.5. The adsorption capacity increased with the increase of phosphate concentration and adsorption temperature. The maximum adsorption capacity of FeOOH@CR material is 3.1 mgPO4 3-/g (62 mgPO4 3-/gFe).The presence of Ca 2+ and Mg 2+ cations enhanced the phosphate adsorption up to 1.21 times, suggesting the use of the material for simultaneous phosphate and hardness removal. In both synthetic and domestic wastewater, the FeOOH@CR was effective for removal phosphate to meet the discharging standards, indicating the practicability of the material in wastewater treatment.