To characterize and confirm the structure of the FZT Y@WDS smart nanoreactors, the following analyzes were performed. The X-ray diffraction (XRD) pattern of the smart nanoreactors from FZT Y@WDS was obtained with the Equinox-3000, Inel France (using a Co-Kα X-ray tube with an input voltage of 40 kV and a scanning speed of 1° (1/min) and 2θ intervals (the value is 0.02°)). The crystalline phase information was confirmed based on the JCPDS database. The XRD patterns for the FZT Y@WDS smart nanoreactors sample are shown in Figure 1A.

It was found that FZT Y@WDS NPs form magnetite with well-defined crystallization in a cubic phase, as shown in Figure 1A. Moreover, it is confirmed that the phase structure is anatase with respect to the second TiO₂ shell, based on the characteristic diffractions of an inverted spinel without impurities for the Fe₃O₄ core (Zhuang et al., 2015) and the monoclinic structure for the first ZrO₂ shell (Heshmatpour and Aghakhanpour, 2011). The Scherrer formula was used to calculate the average crystallite sizes: D = 0.9 × λ β × cos ⁡ θ (2) where θ, diffraction angle; λ is the wavelength of the X-rays; β, FWHM width of the diffraction peak (Klug and Alexander, 1974).

FZT Y@WDS crystals had an average crystallite size of 18.53 nm. In addition to measuring the weight and atomic percentage of the samples, FEI Quanta 200 SEM was used to obtain SEM images, investigate the morphology of the smart structural nanoreactors from FZT Y@WDS, and evaluate the morphology of the smart structural nanoreactors from FZT Y@WDS using EDS or EDAX analysis. Figure 1 shows the results of SEM investigations of nanoparticle size and surface morphology. The microscopic image of the FZT Y@WDS smart structural nanoreactors is shown in Figure 1. According to Figure 1, Fe₃O₄ with a spherical structure was synthesized by the solvothermal method, and zirconia with an almost cubic structure as the first shell on the Fe₃O₄ core, followed by titanium dioxide with a spherical and almost identical structure, can be detected FZT Y@WDS NPs. The size of the particles is almost the same for each nanoparticle and identical in terms of shape. Y@WDS nanoreactors with smart structure for FZT were analyzed by energy dispersive X-ray spectroscopy (EDS) for the presence of O, Fe, Zr, and Ti. The EDAX spectrum shown in Figure 1B confirms the presence of all these elements. According to the results, Ti: Zr: Fe has an atomic ratio of 2.00: 1.5: 1.00. The synthesized FZT Y@WDS NPs showed characteristic peak lines at 0.52 kV for K_α of O and at 0.70, 6.39, and 6.92 kV, which can be assigned to L_α, K_α, and K_β for Fe, respectively. Furthermore, the EDS data of the synthesized sample indicate the presence of zirconium in the synthesized nanoparticles, confirming its presence in the synthesized NPs (characteristic emission lines of 2.04 and 2.26 kV L_α and L_β). Transmission electron microscopy images (TEM) were acquired using a JEOL 2100 microscope at 120 kV. Samples were prepared by suspending the finely powdered products in 100% pure ethanol under sonication and placing a drop of the suspension on a carbon support coated with a gold grid to allow the ethanol to evaporate naturally before analysis. The yolk@shell structure of the Y@WDS NPs shown in Figure 1C was confirmed.

From the particle size distribution (PSD) diagrams in Figure 1D, the PSD of Fe₃O₄, FZ Y@S, FZT Y@DS and FZT Y@WDS NPs are about 30, 60, 140 and 130 nm, respectively, with the average size of FZT Y@WDS smart NPs reduced by about 10 nm due to the wrinkled shell of FZT Y@DS.

Surface analyzers (TriStar-II series, Micromeritics Instrument Corporation, United States) were used to determine the texture properties of the Y@WDS NPs. BET (Bruauer-Emmett-Teller) was used to evaluate the specific surface area (SSA) of the Y@WDS NPs. Based on an isothermal absorption and desorption diagram, we calculated the average pore size and pore size distribution using the Barrett-Joyner-Hacienda (BJH) method. The NPs were first degassed at 300 ^°C for 3 hours before being used in the adsorption-desorption study. The SSA, pore volume, and average pore size of the FZT Y@WDS smart structural nanoreactors were 463 m²/g, 0.46 cm³/g, and 26 nm, respectively. The N₂ gas adsorption-desorption isotherms of FZT Y@WDS structural smart nanoreactors are shown in Figure 2G. IUPAC classification IV type adsorption-desorption isotherm can be used to describe this isotherm. The FZT Y@WDS smart nanoreactors from FZT Y@WDS produced residual H3 (Jung and Park, 2000). The synthesis of FZT Y@WDS smart nanoreactors with increased SSA and larger pore volume, consistent with the observations of SEM and TEM, definitely facilitates the achievement of higher photocatalytic activity under solar irradiation. To measure and determine the chemical composition and bond type on the surface of the FZT Y@WDS sample, XPS analysis was performed using a Thermo-Scientific K-Alpha⁺ model from the United States. Figure 2A shows the chemical composition and surface composition of O, Ti, Fe, and Zr in the synthesized FZT Y@WDS NPs. In order to obtain deep fundamental information about the chemical state between the yolk materials (including C and Fe) and the first and second shells of ZrO₂ and TiO₂, XPS analysis was carried out to search and find the bonding environment of the FZT Y@WDS smart structural nanoreactors. Figures 2D,E,F show the peaks of the energy links associated with Fe, Zr, and Ti, respectively. Carbon-based pollutants are responsible for the presence of carbon in the FZT Y@WDS NPs. According to Figure 2A, the peaks of O1s, Ti2p, and C1s are at 531.33 eV, 461.08 eV, and 288.35 eV, respectively. According to Figure 2F, Ti 2p1/2 and Ti 2p3/2 have peaks at 464.39 eV and 458.69 eV, respectively. The oxidation state Ti⁴⁺ is indicated by the peak between Ti 2p1/2 and Ti 2p3/2. As shown in Figure 2E, the surface spectrum of Zr3d contains double peaks for Zr 3d3/2 and Zr 3d5/2 at binding energies of 188.91 eV and 185.61 eV, respectively. The identification of zirconium as a Zr⁺⁴ oxidation state is indicated by the energy difference of 3.30 eV. A third peak suitable for the shoulder, appearing at the base of Zr 3d3/2, can be attributed to oxygen starvation, which may be due to the misaligned Zr sites of very small ZrO₂ NPs. The strong exchange between Ti and the elements C and Zr can be attributed to the increased binding energy of Ti2p in the FZT Y@WDS smart structure nanoreactors. Figure 2A shows the spectroscopy of the FZT Y@WDS nanoreactors with smart structure. The electron binding energies of O1, Fe2p, Ti2p, C1s, and Zr3d clearly show the presence of the elements O, Fe, Ti, C, and Zr, proving that FZT Y@WDS nanoreactors with smart structures have been successfully synthesized. According to Figure 2C, the peaks of O1s occur at 531.49 eV and 530.43 eV, corresponding to the bulk oxygen band (Ti-O) at the peak point. In addition to the oxygen present in the surface hydroxyl groups, the stronger peak at 531.49 eV is due to the carbonate species. O1 exhibits the same reduction in binding energy as TiO₂ (Figure 2C), again indicating oxygen atomic vacancies in the O1s XPS spectrum. Figure 2B shows the C1s spectrum of the FZT Y@WDS nanoreactors with smart structure found in three charged peaks at 284.89, 286.07, and 288.62 eV. FZT Y@WDS smart structure nanoreactors are used as photon sensitizers to increase the absorption of visible light by residual carbon, causing the peak at 284.89 eV. FZT Y@WDS nanoreactors exhibit two peaks at 288.62 eV and 284.89 eV, which are likely due to adsorption of oxygen-bonded species on the surface. Therefore, the smart structural nanoreactors from FZT Y@WDS harbour multiple carbon species on their surface.

The composition of the photocatalyst can be determined by XPS analysis, but the deviation of the band gap of the heterojunction can also be determined by XPS analysis. VB can be accurately determined by measuring the deviation of the core surfaces before and after contact. The deviation (displacement) of VB is calculated according to Eq. 3 or Eq. 4, so that CB is calculated according to Eq. 5. The exact location of the bandgap structure relative to the NHE is useful for investigation and use in appropriate applications, as mentioned earlier. ∆ E V = E V B M A + E C L A / A B − E C L A − E V B M B + E C L B / A B − E C L B (3) ∆ E V = E C L A / A B − E C L B / A B − E C L A − E V B M A − E C L B − E V B M B (4) ∆ E C = E g A + ∆ E V − E g B (5)

Where ΔE_V is the deflection (displacement) of VB (eV) and ΔE_C is the deflection (displacement) of CB (eV). Figures 2A,B shows the results of XPS analysis and bandgap alignment of FZT Y@WDS. The VB edges of FZ Y@S, FZT Y@DS, and FZT Y@WDS are 2.9, 2.85, and 2.80 eV, respectively, as shown in Figures 2C,D. When Fe₃O₄, ZrO₂, and TiO₂ are combined, changes (shifts) in Ti 2p and Zd 3d core levels occur. When only the DRS and Mott-Schottky analysis is determined and performed, these shifts in the core levels cannot be detected (Figure 2E). These changes (shifts) in the core levels can be used to determine the deflection VB and the deflection CB (Eqs. 4–6) (Figure 2F). A comparison of VB and CB before and after contact shows that VB and CB decrease when Fe₃O₄, ZrO₂, and TiO₂ come into contact. In this case, electrons transferred from ZrO₂ to TiO₂ (band structure increases) or Fe₃O₄ (band structure decreases) could explain the results. The binding energies (531.33 eV, 530.43 eV, and 529.73 eV) are observed with respect to O1s and assigned to the O^2- lattice, the OH⁻ lattice, and adsorbed H₂O molecules, similar to previous studies. This suggests that there is a layer of TiO_2-x (OH)_2x and ZrO_2-x (OH)_2x on the surface of the TiO₂ and ZrO₂ shells.

Determination of the photoluminescence (PL) property of the samples in terms of emission intensity in different wavelengths after their excitation with photons of a certain wavelength was performed by PL analysis or PL spectroscopy (model G9800A from Agilent). PL spectra were measured using a Cary Eclipse fluorescence spectrometer (Agilent Technologies) (at room temperature, the wavelength range of 400–800 nm at an excitation wavelength of 293 nm). Figure 2I shows the analysis of the PL spectrum to calculate the efficiency of charge separation at the sample interface. Figure 2I shows the PL spectrum of FZT Y@WDS. The activation of photocatalysts is strongly influenced by their light absorption. Moreover, photocatalytic activity is closely related to the spectrum and intensity of PL. The highest intensity of all the samples can be seen in Figure 2J. As a result of the yolk@shell structure, increased photon harvesting area, decreased nanoparticle energy band gap, oxygen vacancies, defects in surface states, and other structural impurities in the samples FZ Y@S, FZT Y@DS, and FZT Y@WDS, a peak can be observed in the visible region. The high PL intensity of FZ Y@S clearly indicates that the surface states of ZrO₂ are much lower, and therefore the simple transfer of electrons from VB to CB of ZrO₂ can occur even with a low energy laser excitation source (Ramakrishna et al., 2004). The synergistic effect of FZT Y@WDS nanoreactors with smart structure and the possibility of more (-Fe-O-Zr-), (-Ti-O-Fe-), and (-Ti-O-Zr-) by the PL spectrum of pure catalysts Fe₃O₄ in the core and ZrO₂ in the first shell and TiO₂ in the second shell are physically confirmed (Wu et al., 1984). The results of Figure 2J PL show the peaks of Fe₃O₄ photocatalysts, FZ Y@S, FZT Y@DS and FZT Y@WDS at 543 nm with the above crystallinity. The fluorescence intensity of the Fe₃O₄ composite photocatalyst decreases after the addition of the first ZrO₂ shell. The separation of photoexcited electrons and holes may be enhanced by the ZrO₂ shell’s electronic properties. In addition, ZrO₂ can maintain absorption in the visible range due to its yolk@shell structure and its ability to trap photons. Compared to the smart structural nanoreactors FZT Y@DS, the nanoreactors FZT Y@WDS have significantly improved PL properties. At the excitation wavelength of 360 nm, the ZrO₂ electrons cannot jump from the VB to the CB, so the PL emission of ZrO₂ NPs does not increase upon excitation. However, by reducing the energy band gap with the yolk@shell structure and the presence of voids between the shell and the core of this structure, the problem could be solved. This theory is roughly supported by the waveform shown in Figure 2J. Moreover, coating TiO₂ on FZ Y@S as a second shell significantly increases the surface area and photon emission and harvesting in FZT Y@WDS smart structure nanoreactors, suggesting that TiO₂ as a second shell layer can increase luminescence. While ZrO₂ and TiO₂ alone and due to their chemical inertness prevent the efficient separation of photoexcited carriers. Figure 2K describes the optimal pH (=5.8) for nanoreactors with the FZT Y@WDS smart structure as a function of the type of contaminant and PZC. Catalyst surfaces are neutral at pH = pH_PZC, while negatively and positively charged surfaces are considered at pH>pH_PZC and pH < pH_PZC, respectively. Both adsorption rate and photocatalytic reaction rate are affected by pH, which is obvious. Zeta potential measurements confirmed that stable colloids of Fe₃O₄ NPs, FZ Y@S and FZT Y@WDS were successfully obtained, and Figure 2L shows the zeta potential behavior of these suspensions as a function of pH. Plotting the zeta potential versus pH in Figure 2L shows that the isoelectric points for Fe₃O₄, FZ Y@S and FZT Y@WDS are at pH 4.6, 5.6 and 5.8, respectively. Figure 2L also shows that the surface zeta potential of the FZT Y@WDS smart structural nanoreactors is higher than that of FZ Y@S and Fe₃O₄. The mesoporous coating generates a continuous positive charge in the lower pH range (below the isoelectric point), which could be due to the presence of protonated surface hydroxyl groups (Ti-OH²⁺) and (Zr-OH²⁺). Therefore, as the positive charge density on the adsorbent surface increases at low pH, the zeta potential value increases, while at higher pH, the proton surface sites decrease and the zeta potential becomes more negative. It should be noted that the zeta potential of the particles is positive at low pH, indicating the absorption of H⁺ ions, which can change the surface free energy of ZrO₂ and TiO₂ and create a thermodynamic barrier between anatase and rutile phases. It also reduces the tetragonal and monoclinic phases, allowing martensitic transformation. The high value of the zeta potential for the FZT Y@WDS smart structural nanoreactors in acidic environment indicates the high stability of the NPs in suspension with an initial zeta potential of +30 mV and an isoelectric point (IEP) at pH = 5.8. It is known that electrostatic stabilization is the most common way to stabilize aqueous colloidal suspensions and that the zeta potential of the particle surface is affected by changes on the nanoparticle surface, mainly due to the fact that the oxide surface can be strongly oxidized by hydrogen or hydroxyl ions from the solution, as shown by the reversal of the zeta potential between pH 5.5 and 6.5. However, the interaction between two particles depends on the balance between electrical repulsion and Van der Waals attraction. In aqueous solution, the surface charge of the intelligently structured nanoreactors FZT Y@WDS varies as a function of pH.

Electrochemical analyzes, including cyclic voltammetry and electrochemical impedance spectroscopy, were performed to investigate the effects of the different structures of the samples on performance and electrochemical properties, including charge transfer.

To gain a better understanding of the irregular surface layer of the FZT Y@WDS smart structural nanoreactors, the surface reactivity was investigated by studying the electrocatalytic activity during the production of hydroxyl ions. Figure 3A shows the polarization curves in which the normalized (standard) current density is plotted against the applied potential without iR correction. The Y@WDS FZT sample has a much lower overvoltage (ƞ) (−149.6 mV) at a current density of 15 mA.cm^-2 (vs RHE, as below) than sample Y@DS FZT (−576.6 mV). The corresponding Tafel slope of the FZT Y@WDS sample is 64 mV.dec⁻¹ (Figure 3B), which is lower than that of the FZT Y@DS sample (88 mV.dec⁻¹), indicating higher activity toward hydroxyl ion production. These results indicate that the disordered layer on the surface of FZT Y@WDS can significantly enhance the surface response to hydroxyl ion generation. It is hypothesised that increasing the number of electron carrier trapping sites (i.e., oxygen vacancies) enhances electron transfer, which effectively lowers the energy barrier for proton reduction and activates surface reactivity for hydroxyl ion formation.

The electrochemical properties of Fe₃O₄, FZ Y@S, FZT Y@DS, and FZT Y@WDS samples with a surface area of 1 cm² in contact with 0.5 M sodium sulfate electrolyte were studied by cyclic voltammetry; the resulting plots are shown in Figure 4. As can be seen from Figure 4, structurally coated samples have higher current density than the sample with Fe₃O₄ NPs. The coating in the structure of the nanoreactors and the formation of a heterojunction form a new impurity level above the valence band of the samples, which helps to transfer more electrons, i.e., increase the transfer rate of electrons from the valence band to the conduction band and decrease the recombination rate and electron loss. gives, resulting in more current. With the addition of the coated layer in the designed yolk@shell structure, the rate of charge separation and electron transfer increases and the recombination decreases, which increases the current density.

Electrochemical impedance spectroscopy (EIS) analyzes were performed to integrate the enhanced energy storage properties of the yolk@shell sample. Figure 5 shows the Nyquist diagram of yolk@shell for high (intrinsic) frequencies from the EIS experiments. In the high frequency range, there is a semicircular arc, in the low frequency range a straight line. At low frequencies, the slope of the line is similar, indicating low diffusion resistance of the ions, which is close to the behavior of an ideal capacitor. The ESR can be evaluated as intersecting the real axis of the Nyquist diagram. The ESR of yolk@shell (∼1.2 Ω) is lower than that of other samples (∼1.7 Ω). In the high frequency region, the diameter of the semicircle can be used to measure the charge transfer resistance (R_ct). A comparison of FZT Y@WDS and other samples shows that FZT Y@WDS has a lower R_ct (∼0.1 Ω) than other samples (∼5.4 Ω). The equivalent series resistance (ESR) is connected in parallel with a constant phase element (CPE₁) in parallel with the charge transfer resistance (Rct₁). A constant phase element (CPE₂) is connected in series in parallel with the charge transfer resistance (R_ct2) and a constant phase element (CPE₃) is connected in series in parallel with the charge transfer resistance (R_ct3). To further investigate the resistance of the individual parts of the FZT Y@WDS electrode system of the yolk@shell hollow nanospace structure, the EIS of the yolk@shell hollow nanospace structure of the FZT-Y@WDS electrode was modeled by the modified Randles equivalent circuit (Figure 5C). To investigate the effects of yolk@shell structural architecture on the interfacial properties of FZT Y@WDS electrodes with a surface area of 1 cm², electrochemical impedance spectroscopy analysis was performed in 0.5 M sodium sulfate solution. The data obtained from the experiment were fitted with the mentioned equivalent circuit, and the values of the parameters of the equivalent circuit are listed in Table 1. The mixed level impedance plots (Nyquist, Bode and phase) obtained from the experiment and the fitted plots are shown in Figure 5C,D. Figures 5A,B belong to sample FZT Y@DS and Figures 5C,D also belong to sample FZT Y@WDS (with the same synthesis parameters). The Nyquist diagram of the samples consists of two parts: Charge transfer and mass transfer (diffusion). The curve of the charge transfer part is a more compact semicircle than the perfect semicircle. For this reason, we used a constant phase element instead of an ideal condensor element to model the surfaces of the porous walls, which are heterogeneous and non-ideal. The shape of the Nyquist diagram of the two samples is the same, and the only difference is in the size and magnitude. Comparing the semicircles of the two samples FZT Y@WDS and FZT Y@DS, we find that the semicircle of sample FZT Y@WDS has a smaller diameter. Since the diameter of the semicircle is a measure of the charge transfer resistance, the FZT Y@WDS sample has a lower charge transfer resistance. In other words, the rate of charge transfer and conduction increases and the rate of recombination decreases. Table 1 shows that the FZT Y@WDS sample has a higher CPE-T value than the FZT Y@DS sample (about three times). This shows that the FZT Y@WDS sample has a higher charge conductivity. The resistivity of the FZT Y@DS sample is 1,054 Ω and that of the FZT Y@WDS sample is 100 Ω. Therefore, the charge transfer resistance of FZT Y@DS is greatly reduced due to the wrinkled shell architecture. From the Nyquist, Bode and phase diagrams of the mass transfer (diffusion) section and the data in Table 1, it can be seen that the W-R value of the FZT Y@WDS sample is lower than the W-R value of the FZT Y@DS sample, so it has a lower diffusion resistance. Now we investigate the effects of the nanoreactor structure on the electrochemical properties and the interface of the nanoreactors with EIS. Again, the graphs obtained from the experiment were fitted using the aforementioned equivalent circuit, and the values of the parameters of the equivalent circuit are listed in Table 1; Figures 5C,D show the Nyquist and Bode and phase diagrams, respectively, of sample FZT Y@DS. A comparison of the semicircles of all three samples FZT Y@DS, FZT Y@WDS and FZ Y@S shows that sample FZT Y@WDS has a smaller diameter and thus a lower charge transfer resistance than the other samples. According to Table 1, the charge transfer resistance of FZT Y@DS, FZT Y@WDS and FZ Y@S samples are 147, 100 and 204 Ω / c m 2 , respectively. The FZT Y@WDS sample has higher CPE-T value and lower W-R value than the other samples, which confirms the high conductivity and low penetration resistance of FZT Y@WDS sample.

Adding the second shell and folding the shell increases the void, surface area, and available anatase TiO₂ phase, decreasing the charge transfer resistance. However, if the calcination time is increased, the length of the nanoreactors exceeds the optimal limit and the charge conductivity decreases, i.e., the charge transfer and recombination resistance increases. Therefore, long nanoreactors are not suitable for charge transfer because they slow down the charge transfer.

It can be concluded that the addition of the second shell and the wrinkling of the shell means an increase in the free trapping space, and the size of the surface area is one of the important and influential parameters for the electrochemical properties of FZT Y@WDS electrodes.

Therefore, an optimal length should be selected to transfer the maximum charge. EIS analysis confirms the results of CV analysis.

Electron transfer in heterojunctions is predicted based on the Fermi energy level (E_F), which is an important part of the band structure. VB and E_F of photocatalysts are recorded using UPS. In UPS the horizontal axis intersects the spectral tangent and the intersection is the relative position of VB, when E_F is zero electron volts (0 eV). This approach allows us to visualize the VB, CB, E_F, and trapping modes of a photocatalyst compared to NHE by combining these techniques. In addition, Eq. 7 allows the direct determination of the work function (φ). The work function (φ) (Eq. 8) is equal to the difference between E_vacuum (vacuum energy) (at 4.5 V vs NHE) and the energy function (E_F). To verify that the results are consistent, UPS can be used. The electron transfer occurs between photocatalysts with different work functions (from lower work function to higher work function) when photocatalysts are combined to form heterojunction photocatalysts. Heterojunction photocatalysts irradiated with light can induce internal electric fields that support or prevent further electron transfer. For this reason, there are photocatalysts with S-scheme structures. φ = h v − ∆ E (7) φ = E v a c u u m − E F (8)

UPS the results for Fe₃O₄, ZrO₂ and TiO₂ as well as FZT Y@WDS in Figure 7 show the position of the highest occupied molecular orbital and VB in relation to E_F. By combining the results of DRS and Mott-Schottky analysis, the working power of these materials can be calculated by the difference between the vacuum level and E_F. To compensate E_F, electrons are transferred from a material with a lower work function (ZrO₂) to a material with a higher work function (TiO₂) and (Fe₃O₄). This corresponds to the work function of FZT Y@WDS in Figure 7, and the band alignment of the material before and after contact is shown in Figure 7. The transfer of electrons from ZrO₂ to TiO₂ and Fe₃O₄ bends the band and forms an electric field at the interface of FZT Y@WDS. This is a direct method to validate the latest type of photocatalysts known as S-scheme.

To further investigate the features and characteristics of the separation and transport of light-generated electric charge, transient photocurrent measurements were performed under simulated sunlight and darkness with an on/off pattern for 20 s, as shown in Figure 8A. Pure Fe₃O₄, pure ZrO₂, and pure TiO₂ samples generated current densities of 20 μA.cm^-2, 30 μA.cm^-2, and 25 μA.cm^-2, respectively, under light, while no current was generated in the dark, indicating that the activation of Fe₃O₄, TiO₂, and ZrO₂ by light is confirmed. By forming a heterojunction of pure Fe₃O₄, pure ZrO₂ and pure TiO₂, i.e., FZT Y@DS, the generated luminous current density is 65 μA.cm^-2, which is more than that of pure Fe₃O₄, ZrO₂ and TiO₂. The generated light current density increased to more than 130 μA.cm^-2 due to the wrinkling of the shell in the Y@ WDS FZT structure. This increase in current density is mainly attributed to the efficient generation of electrons (e⁻) and holes (h⁺) by efficient light absorption in FZT Y@WDS, which is confirmed by the results of UV-vis DRS analysis and the efficient electric charge separation due to the formation of heterogeneous bonds. (Heterojunction) between Fe₃O₄, ZrO₂ and TiO₂ is demonstrated. In addition, the presence of voids between the Fe₃O₄ core and the wrinkled shells of ZrO₂ and TiO₂ also contributes to efficient charge separation due to high conductivity and helps in the movement of charges from one semiconductor to another. Such efficient electrical charge separation can lead to higher photocatalytic efficiency. Electrochemical impedance spectroscopy (EIS) is used to further investigate the mechanism of charge transfer at the surface of the synthesized samples. Nyquist EIS curves of FZ Y@S, FZT Y@DS, and FZT Y@WDS were measured in dark and light environments; the results are shown in Figure 8B. The radius of the EIS semicircles corresponds to the total resistance of the electric charge transfer. All samples show lower charge transfer resistance in simulated light than in the dark. FZ Y@S exhibits high charge transfer resistance in light controlled by the semicircle radius, resulting in lower charge separation efficiency. By forming a heterogeneous compound (heterojunction) of Fe₃O₄, ZrO₂ and TiO₂, the charge transfer resistance of FZT Y@DS and FZT Y@WDS is greatly reduced in light, indicating efficient charge transfer by heterojunction. Moreover, the incorporation of pure TiO₂, pure Fe₃O₄ and pure ZrO₂ and the void space between the Fe₃O₄ core and the ZrO₂ and TiO₂ shells, which forms the wrinkled shell FZT Y@WDS, further reduces the charge transfer resistance in light, as shown by the semicircle radius in Figure 8B. From this, it can be seen that the FZT Y@WDS structure operates with a wrinkled shell with a semicircular radius, as shown in Figure 8B. This shows that the empty space in the wrinkled yolk structure of a conductor helps to separate and transfer the photogenerated electric charge (e⁻ and h⁺) from both semiconductors.

By incorporating ZrO₂, TiO₂, and Fe₃O₄ to form FZT Y@WDS, an increase in surface area is observed compared to FZT Y@WDS (see Figure 2G), confirming that the photocatalytic activity of FZT Y@WDS is related to the surface area of the FZT Y@WDS Smart NPs.

The interaction between factors on a given response was studied using three-dimensional diagrams. The middle variable range of such diagrams contains two factors that are changed and the other factors that are constant. NAP degradation is shown in Figure 9 as a function of pH, irradiation time, NAP contaminant concentration, and catalyst dose. The contours in Figure 9A show NAP removal as a function of catalyst loading and NAP pollutant concentration, assuming a constant value for irradiation time of 45 min and pH of 6. The contours in Figure 9B show the NAP removal as a function of pH and NAP pollutant concentration, assuming a constant value for the irradiation time of 45 min and a catalyst loading of 0.3 g/L. Contours in Figure 9C show removal NAP as a function of irradiation time and pollutant concentration NAP, assuming a constant pH of 6 and catalytic loading of 0.3 g/L. The contours plotted in Figure 9D show NAP removal as a function of pH and catalyst loading concentration, assuming a constant value for the irradiation time of 45 min and a NAP contaminant concentration of 20 mg/L. According to Figure 9E, the NAP removal depends on the catalyst loading concentration and the irradiation time, assuming a constant pH of 6 and a pollutant concentration of 20 mg/L. Based on a catalyst loading of 0.3 g/L and a concentration of NAP of 20 mg/L, Figure 9F shows the removal of NAP as a function of irradiation time and pH. Photocatalytic removal of NAP increases with increasing catalyst loading up to 0.5 g/L, as shown in Figure 9. This can be attributed to the enlargement of the active site due to higher charges, which consequently increases the absorption of a higher percentage of photons transferred to the reaction medium and reaction molecules. The photocatalytic removal of NAP increases with decreasing pH of the reaction medium and reaches its maximum value at pH = 3. There is a complex relationship between the pH of the reaction medium and the rate of photocatalytic oxidation. Figure 2 shows that the optimal pH depends on the type of pollutant as well as the PZC, which is 5.8. In the case of NAP, it should be noted that the anion NAP is formed by hydrolysis in its chemical structure. There are two different structures of NAP, which are conceivable at different pH values, both are negatively charged. Naproxen sodium is an organic sodium salt consisting of equimolar amounts of NAP (1-) anions and sodium anions. It should be noted that the pK_a of NAP is equal to 4.15, so that the neutral state is maintained. When the pH is above 5.9 (pH>pH_PZC), the surface of FZT Y@WDS is negatively charged. In the presence of the anion NAP, the surface of the photocatalyst is electrostatically repelled by its negative charge. As a result, the adsorption rate of the NAP anion on the surface of the photocatalyst is reduced, which decreases the decomposition rate. Positive surface charges form on photocatalyst particles below pH_PZC. The molecules of NAP decompose at a pH of about 3.7 (lower than pH_PZC), resulting in a neutral state. Consequently, the adsorption of the NAP anion on the surface of the photocatalyst is enhanced by further lowering the pH to about 3 and by the formation of the NAP anion due to the dissociation of NAP and the formation of a negatively charged anion. This creates a suitable environment for the photocatalytic reaction to occur. As a result of the low pH, •HO₂ radicals are formed form, compensating for low concentrations of •OH radicals and accelerate the degradation reactions. As a result of the low concentration of NAP, the production of OH radicals is more efficient and the recombination reaction is prevented. In addition, the reaction site is less excited, so the reactants have lower diffusion resistance. The combined effect of these two factors is to enhance the photocatalytic reaction and reduce the amount of unreacted NAP in the reaction zone.

During the experimental process, the photocatalytic properties of Fe₃O₄, FZ Y@S and FZT Y@WDS were investigated. The photocatalytic performance of the synthesized NPs was measured for comparison, and the results are shown in Figure 10. As can be seen from Figure 10A,B, the absorption of the solution NAP decreases faster in the presence of nanoreactors with FZ Y@S structure than Fe₃O₄ nanospheres, indicating that the nanoreactors FZ Y@S have a relatively faster degradation rate. In other words, the ZrO₂ shell can improve the photocatalytic performance of the Fe₃O₄ core. When the yolk@shell nanoreactors were used, the degradation rate of NAP was improved (Figures 10B,C), and when the second TiO₂ shell was applied to the surface of the yolk@shell FZ Y@S nanoreactors, the nanoreactors with the yolk@shell FZT Y@WDS structure were synthesized, the photocatalytic efficiency of the nanoreactors with the yolk@shell FZT Y@WDS structure increased dramatically due to the synergistic effect of the ZrO₂ and TiO₂ shells. Photocatalytic degradation of NAP was performed in the presence of FZT Y@WDS nanocomposite materials. Before light irradiation, mixed suspensions of NAP and FZT Y@WDS were magnetically stirred for 30 min to achieve an equilibrium between adsorption and desorption between FZT Y@WDS and NAP. The photodegradation of NAP was monitored by recording the absorption spectrum as a function of the time of light irradiation and is shown in (Figures 10A–C) (Xuan et al., 2009; Pandikumar and Ramaraj, 2012; Jia et al., 2014; Wang et al., 2014; Xu et al., 2015). The absorption intensity of NAP at 575–625 nm decreased with increasing irradiation time. This clearly shows that the concentration of NAP decreased with increasing irradiation time. During light irradiation, absorption of photons and formation of charge separations (e⁻-h⁺) occur due to the excitation of electrons (e⁻) from the valence band of Fe₃O₄, ZrO₂ and TiO₂, leaving holes in the conduction band of TiO₂, ZrO₂ and Fe₃O₄.

The recycling efficiency of catalysts has a great impact on their further use. Therefore, it is necessary to perform a rotation test for the photocatalytic degradation of the solution NAP. As shown in Figure 10D, the FZT Y@WDS nanoreactors still have good photocatalytic activity after eight cycles of degradation of the NAP solution, indicating that the prepared catalysts have excellent stability and ideal catalytic lifetime. In Figure 10D, the separation and reuse of NPs lead to a change in photocatalytic activity. After eight applications of the magnetic photocatalytic particles, the photocatalytic activity decreased slightly (Figure 10D). However, after the regeneration process, in which the catalyst is exposed to UV light in distilled water for 12 h, regeneration restores catalysts to their original state. The activity efficiency of the photocatalyst decreases after several regeneration processes. Deposition of radiation-insensitive materials or loss of NPs during recycling can lead to a reduction in photocatalytic pore activity. The recycling stability of NPs with the designed structure was evaluated using FZT Y@WDS as an example. After each photocatalytic reaction cycle, the photocatalysts were recovered by centrifugation, washed with deionized water, and transferred to a fresh NAP solution to react under the same conditions. As shown in Figure 10E, the photocatalytic activity remains almost unchanged after five consecutive recycling cycles, as the photocatalytic degradation of FZT Y@WDS still exhibits an efficiency of 87% after five cycles. As shown in Figure 10F, XRD analysis of FZT Y@WDS nanoreactors before and after the NAP degradation process confirms the high chemical stability of this nanoreactor.

In this section, the degradation efficiency of NAP from pharmaceutical wastewater was studied, and the results of pH analysis, irradiation time, photocatalyst concentration, and initial concentration of NAP were analyzed. To investigate the effects of nanoreactor composition and structure on photocatalytic performance, photodegradation experiments of NAP under simulated solar radiation (visible light) were performed. Photodegradation of NAP was carried out using the FZT Y@WDS nanoreactor under visible light irradiation (λ> 420 nm) to evaluate its photocatalytic activity. The FZT Y@WDS nanoreactor samples were mixed with a certain amount of NAP solution and stirred in the dark for 0.5 h to reach absorption equilibrium. The change in the concentration of NAP during the photodegradation process is shown in Figure 10G,H. It was found that about 41% of NAP molecules are adsorbed on the surface of the nanoreactor when the absorption reaches equilibrium in the dark, while 21% and 13% of NAP molecules are absorbed on FZ Y@S nanoreactor and Fe₃O₄ NPs, respectively. After 4 h of visible light irradiation, 97% of the compound NAP was almost completely degraded by the FZT Y@WDS nanoreactor, which was higher than the degradation observed with the FZ Y@S nanoreactor (86%) and the Fe₃O₄ nanophotocatalyst (74%). As shown in Figure 10G, the second TiO₂ shell increases the photocatalytic rate of FZT Y@WDS. From Figure 10G, it can be seen that the removal of the impurity NAP has an optimum value because the impurity NAP can also act as a recombination center of light-induced charge carriers and decrease the crystallinity of FZT Y@WDS, FZ Y@S and Fe₃O₄, leading to a decrease in the photocatalytic activity of the photocatalysts. In addition, the Y@WDS has a significant effect on the light-based activity of the catalyst. The natural logarithm of absorption as a function of irradiation time is shown in Figure 10H. It is obvious that the degradation process follows pseudo-first order kinetics according to the research (Kim et al., 2012). As can be seen from the results, the degradation rates by the nanophotocatalysts Fe₃O₄, FZ Y@S, and FZT Y@WDS are 0.588 h⁻¹, 0.846 h⁻¹, and 1.458 h⁻¹, respectively. The contribution of the ZrO₂ and TiO₂ compounds synthesized as shells on the Fe₃O₄ yolk to the degradation rate is 0.258 h⁻¹ and 0.612 h⁻¹, respectively. It can be concluded that both the ZrO₂ and TiO₂ shells significantly enhance the photocatalytic activity of Fe₃O₄ under visible light, and that there is a synergistic effect apparently due to the coexistence of the two compounds ZrO₂ and TiO₂ on the Fe₃O₄ core. The performance of the synthesized photocatalysts with different structural architecture under visible light irradiation determines their degradation rate as follows: