Titanium Isopropoxide (C₁₂H₂₈O₄Ti, 98% Analytical Grade), Urea (CH₄N₂O, 99% Analytical Grade), and Zeolite ((SiO₂) _x (Al₂O₃) _y 86.33% Analytical Grade were bought from Sigma-Aldrich and used as such. In all the experiments, deionized (DI) water was used.

Pour 0.5 g Urea into 25 ml deionized water for the synthesis of TiO₂/Zeolite Nanocomposites (Figure 1). After 5 min of stirring, 5.5 ml of Titanium Isopropoxide was poured dropwise into the solution. The obtained suspension was placed in a water bath at 50°C for 30 min after stirring it for 30 min. Meanwhile, took 50 ml of deionized water and dissolved 4 g of zeolite in it and the resulting suspension was placed in a 37°C water bath for 30 min. The first suspension was poured dropwise into the zeolite suspension, and the final mixture was stirred for 4 h. The final mixture was cooled at room temperature before being aged for 16 h. The product was washed three times with deionized water after filtration. The filtrate was dried for 2–4 h at 60–70°C. The finished product was placed in a vial for future use. Titanium oxide nanoparticles are charged, and because different functional groups are present on the surface of zeolites, nanoparticles can be loaded on the surface of zeolites by functional group and electrostatic interactions.

Scanning Electron Microscope was used to examine the particle size distribution and morphology of the synthesized titanium oxide/zeolite nanocomposites (FEI Nova Nano SEM 450, America). An X-Ray Powder Diffractometer (Bruker AXS D8 Advance) with a high-power radioactive source in the 10–80° range of 2θ was used to examine the crystalline structure of synthesized titanium oxide/zeolite nanocomposites.

Using the following experimental conditions, four batch experiments were performed to investigate the simultaneous adsorption of phosphate and ammonium ions by TiO₂/Zeolite nanocomposites. The initial concentrations (20 mg 100 ml⁻¹), pH (6.00), and temperature (30°C) were fixed in the first experiment. However, the concentration of nanocomposites was varied (0.4, 0.6, 0.7, 0.8, and 1 g 100 ml⁻¹). The concentration of nanocomposites (0.7 g 100 ml⁻¹) was fixed in the second experiment, as was the pH (6.00) and temperature (30°C). While, the initial concentrations of the ions (20, 40, 60, 80, and 100 mg 100 ml⁻¹) were altered. The initial concentrations (20 mg 100 ml⁻¹), nanocomposites concentration (0.7 g 100 ml⁻¹), and pH (6.00) were all fixed in the third experiment. However, the temperature (20, 30, and 40°C) was changed. The starting concentrations of the ions (20 mg 100 ml⁻¹), nanocomposites concentration (0.7 g 100 ml⁻¹), and temperature (30°C) were all fixed in the fourth experiment. However, the pH (4.00, 6.00, 8.00, 10.00, and 12.00) was altered. Final solutions were stirred for 30 min in all four experiments to ensure that the maximum amount of phosphate and ammonium ions were adsorbed by TiO₂/zeolite nanocomposites. The solutions were filtered after shaking for the desired amount of time to remove the adsorbents. A UV-Spectrophotometer (Metash UV-9000S) was utilized for the observation of the adsorption of ions from the solution before and after the addition of TiO₂/zeolite nanocomposites, and all experimental conditions were analyzed to find the best conditions. The efficiency of ion removal from the solution was then found with the help of the following equation: η = C 0   ―   C e C 0 × 100 % (1)

Here η ( % ) is the removal efficiency of phosphorous and nitrogen, C 0 is the starting phosphorous and nitrogen concentration (mg 100 ml⁻¹) and C e is the remaining phosphorous and nitrogen concentration (mg 100 ml⁻¹) after removal (Weng et al., 2013). All experiments were carried out in triplicate, with two parallel samples each.

A small spherical structure of TiO₂ nanoparticles with an average size of 50 nm was revealed by SEM images of TiO₂/zeolite nanocomposites (Figure 2A). It has been suggested that urea can reduce metal ions to produce nanoparticles. Nanoparticles with an amorphous appearance are evenly distributed above zeolites and dispersed in their pore spaces. The zeolite has the appearance of a hollow tube (Figure 2B). The sizes of the nanocomposites vary. Some particles are large, while others are small. The size mainly depends upon the dispersion of nanoparticles on zeolite. Some particles are large due to more dispersion of nanoparticles on zeolite and some particles are small due to less dispersion. Dispersion of the particles on zeolite varies with the synthetic conditions (Monpezat et al., 2019). The aggregation of high-energy surface particles causes large size. The resulting hybrid product had greater potential and advantages for removing ammonium and phosphate ions from water.

The XRD pattern of TiO₂/zeolite nanocomposites revealed a clear, sharp peak, the values were compared to the JCPDS values (Figure 3). JCPDS value confirmed a broad peak at 2θ = 8.04° as a zeolite characteristic peak (Ren et al., 2018). All of the peaks are rougher, indicating that the surface of the zeolite is covered by a stabilizer or capping agent. In this case, nano titanium acts as a stabilizer or capping agent, roughening the zeolite peaks. Specific peaks at 2θ of 13.16°, 35.6°, and 10–35° have appeared and are due to rutile, anatase, and brookite (TiO₂ polymorphs) (Li et al., 2014). The XRD pattern also followed Bragg’s equation (Yuzhen & Shangmin, 1998), indicating that TiO₂/zeolite nanocomposites were formed in all aspects.

Both ammonium (NH4⁺) and phosphate (PO₄ ³⁻) ions were removed with extremely fast kinetics (Figure 4). The removal efficiency for NH₄ ⁺ and PO₄ ³⁻ ions increased to 27.2 and 89.1%, respectively, within the first 5 min. Within 30 min, however, efficiency increased to 38.8% for NH₄ ⁺ ions and 94.6% for PO₄ ³⁻ ions, respectively. Phosphate removal is more efficient and effective than ammonium ion removal. Furthermore, after 15 min, the removal of ammonium ions became constant, whereas the removal of phosphate took 25 min. The adsorption efficiency increased by increasing the time until the equilibrium is reached (Chen et al., 2011). More the time, more will be the chances of agitation of TiO₂ nanoparticles with zeolite (Popuri et al., 2009). But at equilibrium, the surface of the composites is completely blocked by the ions resulting in constant adsorption efficiency. At this point, the adsorption of ions on the surface of the composites becomes equal to the desorption of ions (Dotto & Pinto, 2011). It has been seen that the removal of phosphate ions by these composites is faster. While removal of ammonium ions is also fast but not as efficient as phosphate ions because the removal efficiency also depends upon many factors other than time which is discussed further.

In adsorption experiments, increasing the dose of nanocomposites increased the adsorption efficiency of ammonium ions while decreasing the dose of nanocomposites decreased it. However, the adsorption efficiency of phosphate ions was unaffected (Figure 5A). This nanocomposite has a high capacity for phosphate ions but a low capacity for ammonium ions. As a result, increasing the dose of nanocomposites creates more binding sites for ammonium ions. Adsorption efficiency of ammonium and phosphate ions increased from 2.4 to 52.4 percent and 95.9 to 100 percent. By increasing the dose of the nanocomposites, the available surface area increased. As a result, the interactive absorption sites for both ions were expanded, increasing overall efficiency. However, the static unit volume adsorption capacity and adsorbent dose were reduced. If we keep the contaminant concentration constant, increasing the adsorbent dose will increase the points of contact and thus the adsorbability of the nanocomposites (Chen et al., 2013). The adsorbent activity was reduced at the peak value of the adsorbent dose, resulting in fewer binding sites. As a result, the appropriate dosage of nanocomposites is 0.7 g 100 ml⁻¹.

When the concentrations of the ions were raised from 20 to 100 mg 100 ml⁻¹, the removal efficiency of ammonium ions decreased from 52.4 to 1% and for phosphate ions from 100 to 35.4 percent (Figure 5B). Because of the limited adsorption sites on TiO₂/zeolite nanocomposites, this decrease occurred. In this experiment, the adsorbent dose was held constant while the number of contaminants increased. As a result, competitive adsorption between contaminants occurred (Wan et al., 2017). As a result, adsorption efficiency is reduced. When the initial concentration is lower, more adsorption sites are available to remove contaminants from the nanocomposites’ surface. As a result, maximum adsorption took place. When the initial concentration is higher, however, the minimum adsorption sites are available to remove contaminants from the nanocomposites’ surface. As a result, there was minimal adsorption.

At the optimum temperature, the majority of experiments yield satisfactory results. The removal of ammonium and phosphate ions was studied at three temperatures for the determination of the optimal temperature for ion adsorption. At these three temperatures, the removal efficiencies of ammonium ions were 34.4, 31.0, and 26.7 percent in the first 5 min. However, after 30 min of reaction equilibrium, the ammonium ion removal efficiencies stabilized at 46.3, 37.2, and 33.4 percent, respectively. While phosphate ion removal was high, at 99 percent at all three temperatures, and remained stable (Figure 5C). Because ammonium ion adsorption was an exothermic process, the removal of ammonium ions decreased as the temperature increased. As a result, at higher temperatures, ammonium ion removal was not favored. However, temperature change had not affected the adsorption of phosphate ions. The optimal temperature for the removal of phosphate and ammonium ions from a mixed solution is thought to be 20°C.

At high pH levels, the adsorption efficiency of phosphate and ammonium ions is greatly influenced (Figure 5D). Phosphate adsorption was unaffected until the pH reached 8.00. However, if the pH value raised from 8.00 to 12.00, the phosphate removal decreased from 98.1 to 40.2 percent. The ammonium removal was 52.4 percent at pH 4.00. It stayed at this level until pH 10.00. The removal of ammonium ions dropped to 3.1 percent at pH 12.00, as shown below (Figure 5D). Because of the following reason, the removal efficiency varied with pH. The adsorption sites on the nanocomposites were in a more deprotonated state between pH 4.00 and 10.00 of an aqueous solution, producing a more negative surface charge due to the following mechanism: ≡ T i o H ↔ ≡ T i O − + H + (2)

Ammonium ions can interact more with maximum anion binding sites at pH 4.00 to 10.00. However, at lower pH, more H⁺ ions were produced, which competed with NH4⁺ for adsorption sites and reduced ammonium ion removal efficiency (Huang et al., 2014; Wan et al., 2017). At higher pH, more OH⁻ ions were produced, which competed with PO₄ ³⁻ for adsorption sites and reduced phosphate ion removal efficiency. An optimal pH of 6.00 is required for ion removal.

For a better calculation of adsorption and kinetic isothermal models, different values of removal efficiencies of ammonium and phosphate ions at three temperatures were analyzed for 20–30 h (Figure 6).

Pseudo-first-order and second-order kinetic models were applied for the study of the adsorption mechanism. ln ( q e − q t ) = ln q e − k 1 t (3) t q t = 1 k 2 q e 2 + 1 q e (4) h = k 2 q e 2 (5)

Where q_t and q_e are adsorption capacities at the time (min) and equilibrium and k₁ (min⁻¹) and k₂ (g mg⁻¹ h⁻¹) are the pseudo-first-order adsorption and pseudo-second-order adsorption rate constants, respectively. Table 1 shows the proper values for all parameters.

The pseudo-first-order correlation coefficient (R2) values for NH4⁺ and PO₄ ³⁻ are 0.941, 0.953, 0.878 and 0.553, 0.592, 0.862, respectively. While the pseudo-second-order values for NH₄ ⁺ and PO₄ ³⁻ are 0.987, 0.966, 0.974 and 1.000, 1.000, 1.000, respectively. As a result, it was determined that pseudo-second order is a more consistent kinetic model for hybrids’ removal behavior. Both the correlation coefficient (R₂) values for NH4⁺ and PO₄ ³⁻ ions at three temperatures are greater than 0.990, indicating that these ions are removed entirely by chemical adsorption. Values of correlation coefficient at lower temperature are greater than 0.990 suggested that the adsorption process was not first order reaction but followed the pseudo second order kinetic model (Wu et al., 2009). The adsorption process took place in many steps and the rate determining step involved the sharing or exchange of electrons suggesting that the adsorption process was mainly chemisorption. The adsorption of the ions followed an exothermic reaction and should be taken place at lower temperature for the maximum adsorption of the ions.

Thus, high values of correlation coefficient for ions are due to the pseudo second order kinetics and chemisorption exothermic process (Ho & McKay, 1999). As a result of the interaction between ions and nanocomposites, good removal efficiency of both NH₄ ⁺ and PO₄ ³ ions was observed. This interaction occurred via ligand exchange or electron sharing between NH₄ ⁺—PO₄ ³⁻ ions and nanocomposites. The rate constants k₂ of NH₄ ⁺ ions for pseudo-second-order remained unaffected by increasing the temperature, but PO₄ ^3- ions varied more efficiently. It was discovered that this method is faster, more practical, and simple for controlling pollution caused by these ions. The temperature had not affected the removal of NH₄ ⁺ ions with nanocomposites; it demonstrated that NH4⁺ ion adsorption with nanocomposites remained stable at these three temperatures. While increasing the temperature reduced PO₄ ^3- ion adsorption with nanocomposites, it revealed that PO₄ ³⁻ ion adsorption with nanocomposites is an exothermic process. At a lower temperature, it would remain stable.

The adsorption data were also fitted into Freundlich and Langmuir equations (Table 2) to determine the precise adsorption capacity of the nanocomposites for these two water pollutants (Zhang et al., 2011). Eq. 6 depicts the Freundlich model equation: ln q e = ln K F − 1 n ln C e (6)

While q_e is called equilibrium adsorption capacity (mg g⁻¹), K_F is known as the Freundlich equilibrium constant (L g⁻¹), 1/n is constant for adsorption capacity, and C_e is NH₄ ⁺ and PO₄ ³⁻ equilibrium concentration (mg L⁻¹).

While Eq. 7 is used to express the Langmuir isotherm model: C e q e = C e q 0 + 1 q 0 K L (7)

K_L (L mg⁻¹) is called the Langmuir adsorption constant and q_m (mg g⁻¹) is known as the maximum adsorption capacity.

Based on correlation coefficient (R ²) values for the Freundlich & Langmuir models, ammonium and phosphate ions adsorb on nanocomposites via a monolayer adsorption process. Because ammonium and phosphate ions have a positive and a negative charge, NH₄ ⁺ is less adsorbed on nanocomposites than PO₄ ³⁻. Adsorption is aided by negative charges. The 1/n values of ammonium and phosphate ions are both very low, indicating that the adsorption process was mainly chemical.

K_d (Thermodynamic distribution coefficient), ΔG° (kJ mol⁻¹ Gibbs free energy), ΔH° (kJ mol⁻¹ Enthalpy change) and ΔS° (J mol⁻¹ K⁻¹ Entropy change) were used for the calculation of thermodynamic values by using Eqs 8, 9: ln K d = − Δ G ° R T = − Δ H ° R T + Δ S ° R (8) Δ G ° = ‐R T ⁡ ln K d (9)

When a graph was plotted between ln K_c and 1/T, ΔS°, Δ,H°, and ΔG° parameters were found from the intercept and slope of the graph (Table 3).

The ΔH° and ΔS° values for ammonium ion adsorption confirmed that the process of adsorption of these ions was exothermic. ΔG° negative values for ammonium ion adsorption confirmed that this was a highly favorable and spontaneous process. Phosphate ion adsorption ΔH°, ΔS°, and ΔG° values confirmed that the process was exothermic and spontaneous. ΔG° values decreased as temperature increased, confirming that higher temperatures were unfavorable for the removal process. The above calculations and conclusions confirmed that the removal of NH₄ ⁺ by TiO₂/zeolite nanocomposites was a physical adsorption process because the surface of the nanocomposites has a negative charge in water that electrostatically attracts NH₄ ⁺ ions. While phosphate adsorption was caused by the addition of nanocomposites to water, Ti-OH bonds form as a result of hydrolysis at the titanium surface. TiPO₄ ²⁻ is formed when this bond exchanges with phosphate.

TiO₂/zeolite nanocomposites absorbed fewer ammonium ions (3.75 mg g⁻¹) than natural zeolite (17.68 mg g⁻¹), modified zeolite combined with struvite crystallization (17.55 mg g⁻¹), and iron nanoparticles (9.70 mg g⁻¹), but more than eucalyptus magnetic nanocomposites (3.47 mg g⁻¹). Phosphate adsorption by TiO₂/zeolite nanocomposites (38.63 mg g⁻¹) was greater than that of natural zeolite (2.73 mg g⁻¹) and modified zeolite combined with struvite crystallization (8.25 mg g⁻¹), but it was slightly lower or nearly equal to that of eucalyptus magnetic nanoparticles dispersed on zeolite (38.91 mg g⁻¹) (Huang et al., 2014; Wang et al., 2014; Wan et al., 2017; Xu et al., 2020). The prepared TiO₂/zeolite nanocomposites have selective adsorption for NH₄ ⁺ and PO₄ ³⁻, with PO₄ ³⁻adsorption being greater than NH₄ ⁺ adsorption.