Synthesis of Gd 2 O 3 Nanoparticles and Their Photocatalytic Activity for Degradation of Azo Dyes

: Gadolinium oxide (Gd 2 O 3 ) nanoparticles were prepared via the reaction of gadolinium nitrate hexahydrate (Gd (NO 3 ) 3 · 6H 2 O) and ethylamine (C 2 H 5 NH 2 ), and their surface morphology, particle size, and properties were examined by using scanning electron microscopy, X-ray diffraction (XRD), Raman spectroscopy, Fourier transform infrared (FT-IR) spectroscopy, and ultraviolet visible (UV-vis) spectroscopy. The Gd 2 O 3 nanoparticles were used as the photocatalyst for the degradation of various azo dyes, such as methyl orange (MO), acid orange 7 (AO7), and acid yellow 23 (AY23) under irradiation with UV light. The effect of the experimental parameters (initial concentration of azo dyes, dosage of catalyst, and wavelength of UV light) on the photocatalytic properties of the Gd 2 O 3 nanoparticles were investigated. At a constant H 2 O 2 concentration, the photocatalytic degradation efﬁciency of the Gd 2 O 3 nanoparticles for various azo dyes was in the order: methyl orange > acid orange 7 > acid yellow 23. The kinetics study showed that the photocatalytic degradation of azo dyes was followed by a pseudo ﬁrst-order reaction rate law.


Introduction
The use of a metal oxide semiconductor photocatalyst is a common technique for the degradation of organic pollutants [1,2]. The textile industry generates a lot of wastewater, which contains a lot of dyes. Many contaminants in the ecosystem have been problematic due to their stable chemical structure, posing various teratogenic, carcinogenic, and mutagenic risks [3]. As a result, wastewater purification has always received a lot of interest in the scientific field. The combined advanced oxidation process (AOP), which includes heterogeneous photocatalysis and hydrogen peroxide, provides the synergistic effect for degradation of organic dyes in wastewater [3][4][5][6][7]. The dye degradation mechanism of the AOP involves the generation of hydroxyl radicals (·OH) and superoxide anion radicals (·O 2 − ) [8][9][10]. Their radicals are unstable and can attack organic pollutants to make harmless products.
Lanthanide compounds have been used in photocatalysis, owing to their unique f -electronic configuration and bandgap energy. Rare earth elements have half-filled 4fshells with unpaired electrons and often exhibit an empty 5d shell. They are used in a wide range of applications, such as fluorescent materials, high-resolution X-ray medical imaging, ultraviolet (UV) detectors, catalysts, and dopants [11][12][13][14][15][16]. Owing to its high chemical stability, UV absorption, and active photon-to-electron conversion, gadolinium oxide (Gd 2 O 3 ) is widely used as an n-type semiconductor photocatalyst [12,17,18].
Recently, several papers have been reported on photocatalytic degradation by using lanthanide oxide nanoparticles [11,12,15,16]. However, there are still insufficient studies available on the degradation of lanthanide oxide nanoparticles of organic pollutants. Gd 2 O 3 nanoparticles can be used for the photocatalytic degradation of organic pollutants and wastewater treatment, and, due to their high photocatalytic activity, they are also used as an additional oxidizing agent along with other agents, such as hydrogen peroxide [19][20][21].
In Gd 2 O 3 , the electrons in the valence band move to the conduction band under UV irradiation [22]. Gadolinium oxide nanoparticles effectively produce free radicals and their recombination with holes is avoided because of the presence of half-filled 4f orbitals and an empty 5d shell, which often serve as trapping centers to prevent the combination of carriers [23]. The ·OH radicals generated in water under UV irradiation are responsible for the degradation of azo dyes [19]. The unique 4f orbitals/electrons of rare earth compounds have magnetic, electrical, optical, phosphorus, and catalytic properties [17,24].
AOPs are based on the combination of strong oxidants with a catalyst and radiation; the addition of hydrogen peroxide increases the rate [25]. The photocatalytic degradation efficiency of organic dyes may be enhanced in combination with H 2 O 2 and Gd 2 O 3 nanoparticles under UV irradiation. ln this study, we carried out photocatalytic degradation of various organic dyes, such as methyl orange (MO), acid orange 7 (AO7), and acid yellow 23 (AY23) under different conditions using the Gd 2 O 3 nanoparticles.

Characterization of Gd 2 O 3 Nanoparticles
For Gd 2 O 3 , the XRD peaks observed at 2θ = 28.64 • correspond to the (222) plane of the cubic phase (JCPDS 3-065-3181) [24], and those at 2θ = 20.14 • , 33.16 • , 47.64 • , and 56.53 • correspond to the (211), (400), (440), and (622) planes, respectively, of the hexagonal phase ( Figure 1). The size of the synthesized Gd 2 O 3 nanoparticles was calculated using the Scherrer formula: where D is the size of the crystallites, K is the Scherrer constant, λ is the X-ray wavelength, β is the full width at half maximum (in radians) of the XRD peak, and θ is the Bragg angle. The average crystallite size of the Gd 2 O 3 nanoparticles was calculated to be 13.7 nm [24].
Recently, several papers have been reported on photocatalytic degradation by using lanthanide oxide nanoparticles [11,12,15,16]. However, there are still insufficient studies available on the degradation of lanthanide oxide nanoparticles of organic pollutants.
Gd2O3 nanoparticles can be used for the photocatalytic degradation of organic pollutants and wastewater treatment, and, due to their high photocatalytic activity, they are also used as an additional oxidizing agent along with other agents, such as hydrogen peroxide [19][20][21].
ln Gd2O3, the electrons in the valence band move to the conduction band under UV irradiation [22]. Gadolinium oxide nanoparticles effectively produce free radicals and their recombination with holes is avoided because of the presence of half-filled 4f orbitals and an empty 5d shell, which often serve as trapping centers to prevent the combination of carriers [23]. The ·OH radicals generated in water under UV irradiation are responsible for the degradation of azo dyes [19]. The unique 4f orbitals/electrons of rare earth compounds have magnetic, electrical, optical, phosphorus, and catalytic properties [17,24].
AOPs are based on the combination of strong oxidants with a catalyst and radiation; the addition of hydrogen peroxide increases the rate [25]. The photocatalytic degradation efficiency of organic dyes may be enhanced in combination with H2O2 and Gd2O3 nanoparticles under UV irradiation. ln this study, we carried out photocatalytic degradation of various organic dyes, such as methyl orange (MO), acid orange 7 (AO7), and acid yellow 23 (AY23) under different conditions using the Gd2O3 nanoparticles.

Characterization of Gd2O3 Nanoparticles
For Gd2O3, the XRD peaks observed at 2θ = 28.64° correspond to the (222) plane of the cubic phase (JCPDS 3-065-3181) [24], and those at 2θ = 20.14°, 33.16°, 47.64°, and 56.53° correspond to the (211), (400), (440), and (622) planes, respectively, of the hexagonal phase ( Figure 1). The size of the synthesized Gd2O3 nanoparticles was calculated using the Scherrer formula: where D is the size of the crystallites, K is the Scherrer constant, is the X-ray wavelength, is the full width at half maximum (in radians) of the XRD peak, and is the Bragg angle. The average crystallite size of the Gd2O3 nanoparticles was calculated to be 13.7 nm [24]. 10   XPS analysis determined the surface elemental composition and their ionic states. Figure 2a depicts the appearance of Gd 3d, Gd 4d, and O 1s elements of the Gd2O3 nano- XPS analysis determined the surface elemental composition and their ionic states. Figure 2a depicts the appearance of Gd 3d, Gd 4d, and O 1s elements of the Gd 2 O 3 nanoparticles in the XPS survey spectrum. The binding energy at 1187.9 and 1219.3 eV corresponds to the Gd 3d 5/2 and Gd 3d 3/2 , respectively, in Figure 2b [12,18]. Figure 2c demonstrates that the binding energy at 142.2 eV and 147.8 eV were ascribed to the Gd 4d 5/2 and Gd 4d 3/2 spin-orbits, respectively. As for Figure 2d, the O 1s peak in the Gd 2 O 3 nanoparticles displayed the binding energy at 531.7 eV [19]. as shown in Figure 3e [15,19].
The vibrational modes of Gd2O3 nanoparticles were observed by Raman spectroscopy in Figure 4a. In the Raman spectrum, three peaks were observed: two at 313 and 443 cm −1 , corresponding to the Fg + Eg modes and the Fg mode, respectively, and most importantly, the prominent peak at 358 cm −1 is assigned to the Fg + Ag modes, which mainly corresponds to the cubic phase of the Gd2O3 nanoparticles [26]. (c) (d) The Gd2O3 nanoparticles were investigated by FT-IR spectroscopy. The Gd2O3 nanoparticles show the characteristic peak at 547 cm −1 due to the banding vibration of Gd-O in the FT-IR spectroscopy in Figure 4b [23,24]. The SEM image of Gd 2 O 3 in Figure 3a have a pine needle-like shape. The TEM image of Gd 2 O 3 in Figure 3b shows a rod shape with the length of 100−900 nm. Additionally, in Figure 3c,d, the elemental compositions of the synthesized compounds were determined by EDX mapping/elemental analysis. According to the EDX spectrum, the Gd 2 O 3 nanoparticles contained a gadolinium atom (Gd) of 34.31%, and an oxygen atom (O) of 65.69%, as shown in Figure 3e [15,19].
The vibrational modes of Gd 2 O 3 nanoparticles were observed by Raman spectroscopy in Figure 4a. In the Raman spectrum, three peaks were observed: two at 313 and 443 cm −1 , corresponding to the F g + E g modes and the F g mode, respectively, and most importantly, the prominent peak at 358 cm −1 is assigned to the F g + A g modes, which mainly corresponds to the cubic phase of the Gd 2 O 3 nanoparticles [26].
teristic constant of the semiconductor, Eg is the apparent optical band gap of the material, and k is a constant associated with electronic transition types (for direct allowed transition: k = 1/2, for direct forbidden transitions: k =2/3, for indirect allowed transition: k =2, and for indirect forbidden transitions: k = 3). According to the literature [23,27], gadolinium oxide nanoparticles were characterized by a direct allowed electronic transition, so k = 1/2 was used. The value of Eg was evaluated by extrapolating the linear portion of the curve in Figure 5b. This was found to be 5.04 eV for the Gd2O3 nanoparticles.    When aqueous solutions of azo dyes were irradiated by UV light without adding a catalyst, the azo dye degradation process was very slow. The addition of the Gd2O3 nanoparticles, i.e., the catalyst, accelerated the azo dye degradation process.
The aqueous solution of MO containing the H2O2 and Gd2O3 nanoparticles showed an absorbance peak at 480 nm ( Figure 6a). During the photocatalytic degradation reaction, no other absorbance peak was observed. The absorbance peak in the visible region faded away after UV irradiation at 254 nm for 100 min. The absorption peak appeared at 480 nm, and the intensity of this peak decreased with UV light irradiation. With time, this peak shifted to a lower absorbance in the visible region. The MO showed a degradation  The absorbance peak of the Gd 2 O 3 nanoparticles was observed at 228 nm in UV-vis spectroscopy [27]. The optical bandgap energy (E g ) was estimated by the method proposed by Tauc [17]. The optical band gap of Gd 2 O 3 is determined by the following equation: where α is absorption coefficient, h is Plank constant, ν is light frequency, A is a characteristic constant of the semiconductor, E g is the apparent optical band gap of the material, and k is a constant associated with electronic transition types (for direct allowed transition: k = 1/2, for direct forbidden transitions: k =2/3, for indirect allowed transition: k =2, and for indirect forbidden transitions: k = 3). According to the literature [23,27], gadolinium oxide nanoparticles were characterized by a direct allowed electronic transition, so k = 1/2 was used. The value of E g was evaluated by extrapolating the linear portion of the curve in Figure 5b. This was found to be 5.04 eV for the Gd 2 O 3 nanoparticles. The histograms of photocatalytic degradation efficiency (MO, AO7, AY23) under various conditions (UV/H2O2, H2O2/Gd2O3, UV/H2O2/Gd2O3, and UV/Gd2O3 for 100 min) are shown in Figure 7. It can be observed from the spectra that the Gd2O3 nanoparticles accelerated the degradation of the azo dyes. The photocatalytic degradation efficiency was the highest when both Gd2O3 and H2O2 were used under UV irradiation conditions. When aqueous solutions of azo dyes were irradiated by UV light without adding a catalyst, the azo dye degradation process was very slow. The addition of the Gd 2 O 3 nanoparticles, i.e., the catalyst, accelerated the azo dye degradation process.
The aqueous solution of MO containing the H 2 O 2 and Gd 2 O 3 nanoparticles showed an absorbance peak at 480 nm ( Figure 6a). During the photocatalytic degradation reaction, no other absorbance peak was observed. The absorbance peak in the visible region faded away after UV irradiation at 254 nm for 100 min. The absorption peak appeared at 480 nm, and the intensity of this peak decreased with UV light irradiation. With time, this peak shifted to a lower absorbance in the visible region. The MO showed a degradation of 79.68% ( Figure S1c) and 26.62% ( Figure S1a) in the presence and absence of the catalyst, respectively. The histograms of photocatalytic degradation efficiency (MO, AO7, AY23) under various conditions (UV/H2O2, H2O2/Gd2O3, UV/H2O2/Gd2O3, and UV/Gd2O3 for 100 min) are shown in Figure 7. It can be observed from the spectra that the Gd2O3 nanoparticles accelerated the degradation of the azo dyes. The photocatalytic degradation efficiency was the highest when both Gd2O3 and H2O2 were used under UV irradiation conditions.  The absorbance peak of AO7 appeared at 485 nm ( Figure 6b). After 100 min of UV irradiation in the absence of the catalyst, only 6.11% ( Figure S2a) of the AO7 dye was degraded. On the other hand, a 79.92% ( Figure S2c) degradation was achieved in the presence of the Gd 2 O 3 catalyst.
AY23 showed an absorption peak at 428 nm ( Figure 6c). The AY23 was the active site for an oxidative attack. The AY23 showed a degradation of 59.14% ( Figure S3c) and 3.81% ( Figure S3a) in the presence and absence of Gd 2 O 3 catalyst, respectively.
The adsorption of H 2 O 2 onto Gd 2 O 3 particles can modify their surface [28][29][30]. The dye adsorption capacity on metal oxides is an important parameter for surface reactions; it depends on the surface area and other surface-related parameters, such as the surface density, pore size distribution, and morphology of the particle surface. If the adsorption of dye molecules increases, the rate of photocatalytic degradation increases. The photocatalytic degradation efficiency of semiconductor oxides under UV irradiation is related to the electronic band structure of the semiconductor.   The photocatalytic degradation efficiency of azo dyes using Gd 2 O 3 is calculated by the following equation: where C 0 is the concentration of azo dye at the adsorption-desorption equilibrium in the dark (t = 0) and C t is the concentration of azo dye at reaction time t (min).

Mechanism of Photocatalytic Degradation of Azo Dyes by Gd 2 O 3 Nanoparticles
Upon UV irradiation, the electrons in the valence band of the catalyst moved into the conduction band. This resulted in the continuous generation of holes (h + ) in the valence band and electrons (e − ) in the conduction band. The generation of electron-hole pairs contributed to the activity of the photocatalyst. The holes with a high oxidative ability can oxidize OH − to ·OH radicals. The holes in the valence band and electrons in the conduction band were generated and combined with a hydroxide ion and oxygen to generate ·OH and ·O 2− radicals. The photocatalytic degradation of the organic dyes occurred on the surface of the catalyst. The ·OH and ·O 2 − radicals act as oxidizing agents for azo dye molecules [19].
The azo dye degradation mechanism under the UV/H 2 O 2 /Gd 2 O 3 condition is as follows (Figure 8) [12,15,19]: Azo dye + ·OH → Products of photocatalytic degradation (10) oxidize OH − to ·OH radicals. The holes in the valence band and electrons in the conduction band were generated and combined with a hydroxide ion and oxygen to generate ·OH and ·O 2− radicals. The photocatalytic degradation of the organic dyes occurred on the surface of the catalyst. The ·OH and ·O2 − radicals act as oxidizing agents for azo dye molecules [19]. The azo dye degradation mechanism under the UV/H2O2/Gd2O3 condition is as follows ( Figure 8) [12,15,19]: Gd2O3(hvb + ) + (H2O) → ·OH + H − Azo dye + ·OH → Products of photocatalytic degradation (10) The synergetic behavior is understandable considering the electron-capturing role of the hydrogen peroxide as an external electron hole recombined with the gadolinium oxide surface. However, it can also improve the hydroxyl radical introduction in the reaction [6,31].  The synergetic behavior is understandable considering the electron-capturing role of the hydrogen peroxide as an external electron hole recombined with the gadolinium oxide surface. However, it can also improve the hydroxyl radical introduction in the reaction [6,31].

Kinetics Study for Photocatalytic Degradation of Azo Dyes
In a first-order reaction, the rate (ν) of the reaction the rate is expressed as follows: where c is the concentration of reactant A, k is the first-order rate constant, and t is time.
The rate depends on the concentration of the reactant A present at time, t. In photocatalytic degradation of azo dyes, the integrated form of rate Equation (12) may be written as in Equation (16). Separating variables of Equation (12) and taking an integral calculus, − ln C = kt + Integration constant (14) The value of the integration constant is determined when time is zero (t = 0) of the following expression: Catalysts 2021, 11, 742 9 of 18 Therefore, by Equations (14) and (15), we can obtain Equation (16) [32].
ln C C 0 = −kt (16) where C 0 is the concentration of the reactant (azo dye) at an initial time, t = 0, C is the concentration of the reactant (azo dye) at a specific time t, and k is the first-order rate constant [33]. The linear behavior of the curves confirms that the photocatalytic degradation of azo dyes, which were MO, AO7, and AY23, followed a pseudo first-order reaction rate law. As it can be observed from Figure 9, the values of R 2 (coefficient of determination) for the pseudo first-order reaction kinetics were (a) 0.99888, (b) 0.99867, (c) 0.99555.
The value of the integration constant is determined when time is zero (t = 0) of the following expression: −ln = Integration constant (at t = 0) (15) Therefore, by Equations (14) and (15), we can obtain Equation (16) [32]. ln = −kt (16) where C0 is the concentration of the reactant (azo dye) at an initial time, t = 0, C is the concentration of the reactant (azo dye) at a specific time t, and k is the first-order rate constant [33]. The linear behavior of the curves confirms that the photocatalytic degradation of azo dyes, which were MO, AO7, and AY23, followed a pseudo first-order reaction rate law. As it can be observed from Figure 9, the values of R 2 (coefficient of determination) for the pseudo first-order reaction kinetics were (a) 0.99888, (b) 0.99867, (c) 0.99555.

Effect of Gd2O3 Nanoparticle Dose for Photocatalytic Degradation of Azo Dyes
The effect of the Gd2O3 dose on the photocatalytic degradation of azo dyes was investigated. An increase in the dose of the catalyst increased the photocatalytic efficiency and generated a large number of electron-hole pairs. A large number of hydroxyl radicals were available for azo dye degradation. Under the same conditions, the photocatalytic degradation of MO by Gd2O3 nanoparticles improved the reaction rate constant from

Effect of Gd 2 O 3 Nanoparticle Dose for Photocatalytic Degradation of Azo Dyes
The effect of the Gd 2 O 3 dose on the photocatalytic degradation of azo dyes was investigated. An increase in the dose of the catalyst increased the photocatalytic efficiency and generated a large number of electron-hole pairs. A large number of hydroxyl radicals were available for azo dye degradation. Under the same conditions, the photocatalytic degradation of MO by Gd 2 O 3 nanoparticles improved the reaction rate constant from −0.00379 to −0.02855 min −1 by increasing the catalyst dose from 0 to 1.5 g/L after a 100 min degradation reaction. Under the same conditions, the photocatalytic degradation of AO7 by Gd 2 O 3 nanoparticles improved the reaction rate constant from −0.00060 to −0.02454 min −1 by increasing the catalyst dose from 0 to 1.5 g/L after a 100 min degradation reaction. The photocatalytic degradation of AY23 by Gd 2 O 3 nanoparticles improved the reaction rate constant from −0.00030 to −0.03712 min −1 by increasing the catalyst dose from 0 to 1.5 g/L after a 100 min degradation reaction. The photocatalytic degradation rate and removal% of the aqueous azo dye solution increased when the catalyst dose also increased. Under the same conditions, the degradation rate of azo dyes increased when the adsorption of the dye molecules on the catalyst surface was increased ( Figure 10, Table 1).

Effect of Gd2O3 Nanoparticle Dose for Photocatalytic Degradation of Azo Dyes
The effect of the Gd2O3 dose on the photocatalytic degradation of azo dyes was investigated. An increase in the dose of the catalyst increased the photocatalytic efficiency and generated a large number of electron-hole pairs. A large number of hydroxyl radicals were available for azo dye degradation. Under the same conditions, the photocatalytic degradation of MO by Gd2O3 nanoparticles improved the reaction rate constant from −0.00379 to −0.02855 min −1 by increasing the catalyst dose from 0 to 1.5 g/L after a 100 min degradation reaction. Under the same conditions, the photocatalytic degradation of AO7 by Gd2O3 nanoparticles improved the reaction rate constant from −0.00060 to −0.02454 min −1 by increasing the catalyst dose from 0 to 1.5 g/L after a 100 min degradation reaction. The photocatalytic degradation of AY23 by Gd2O3 nanoparticles improved the reaction rate constant from −0.00030 to −0.03712 min −1 by increasing the catalyst dose from 0 to 1.5 g/L after a 100 min degradation reaction. The photocatalytic degradation rate and re-moval% of the aqueous azo dye solution increased when the catalyst dose also increased. Under the same conditions, the degradation rate of azo dyes increased when the adsorption of the dye molecules on the catalyst surface was increased ( Figure 10, Table 1).

Effect of Initial Dye Concentration on the Photocatalytic Degradation of Azo Dyes
The effect of the initial azo dye concentration on the photocatalytic degradation of azo dyes was investigated. The photocatalytic degradation experiments were carried out at different azo dye concentrations and at a constant H 2 O 2 concentration. The initial azo dye concentration varied from 0.042 to 0.126 mM. At high initial dye concentrations, the path length of the photons entering the solution decreased and the number of ·OH radicals generated in the solution also decreased ( Figure 11, Table 2).

Effect of UV Light Wavelength on Azo Dye for Photocatalytic Degradation
In the presence of the Gd 2 O 3 photocatalyst, UV irradiation at 365 nm (UV-A) caused higher degradation of the azo dyes than that at 254 nm (UV-C) with reduced energy consumption. For the same amount of photocatalyst, the photocatalytic degradation efficiency of MO, AO7, and AY23 obtained with UV-A radiation was higher than that obtained with UV-C [25,34,35]. This can be attributed to the higher generation rate of ·OH radicals and, consequently, the higher reaction rates at 365 nm [36]. UV-A photons penetrate much deeper into the solution and degrade a large number of dye molecules on the surface of Gd 2 O 3 [37]. It was found that the percentage of absorbance for UV light increased significantly in the UV-A and visible regions, which favored the photocatalytic performance of the catalyst (Figure 12, Table 3) [38]. dation of MO dye solution for each cycle, but no apparent Gd2O3 photocatalyst deactivation after five cycles [2,18]. This is because, after the photocatalytic degradation of MO dye, the photocatalyst was rinsed with distilled water, but a little bit of the MO dye remained on the photocatalyst's surface. As a result, the reusability of photocatalyst efficiency for the MO dye degradation was reduced, as shown in Figure 13.

Reusability of Gd 2 O 3 Nanoperticles as Photocatalyst for Degradation of MO
The recyclability of the photocatalyst is a significant factor in its industrial application [39,40]. The recyclability of the Gd 2 O 3 nanoparticles was studied by repeating the experiment five times for the photocatalytic degradation of MO solution ( Figure 13). The recyclability of the Gd 2 O 3 photocatalyst demonstrates a little decrease in the photodegradation of MO dye solution for each cycle, but no apparent Gd 2 O 3 photocatalyst deactivation after five cycles [2,18]. This is because, after the photocatalytic degradation of MO dye, the photocatalyst was rinsed with distilled water, but a little bit of the MO dye remained on the photocatalyst's surface. As a result, the reusability of photocatalyst efficiency for the MO dye degradation was reduced, as shown in Figure 13.
particles for the degradation of the azo dyes MO, AO7, AY23 in the presence of H2O2 at diff UV wavelengths.

Azo Dye
Wavelength of UV Lamp (nm)

Methods
The photocatalytic degradation of the azo dyes was carried out using a UV lamp (4 W, 254 nm; 4 W, 365 nm) and was confirmed by ultraviolet-visible (UV-Vis) spectroscopy (Shimazu UV-1601 PC, Tokyo, Japan). The morphology of the catalyst was observed by using scanning electron microscopy (SEM, JEOL Ltd., JSM-6510, Tokyo, Japan) at an acceleration voltage of 10 kV. The vibrational state of the catalyst was investigated by Raman spectroscopy at 532 nm wavelength (BWTEK i-Raman Plus, Newark, DE, USA). Fourier-transform infrared (FT-IR) spectroscopy (Thermo Scientific Nicolet iS10, Madison, WI USA) was used to obtain information on the functional groups in the catalyst. The surface elemental composition and their ionic states was analyzed using X-ray photoelectron spectroscopy (XPS) (NEXSA, Thermo Fisher Scientific USA). The crystal structures and average crystallite size of the catalysts were analyzed using powder XRD (Bruker, D8 Advance, Karlsruhe, Germany). The morphology and elemental composition of the catalyst were measured using a transmission electron microscope (TEM) (NEO ARM, JEOL, Japan) with energy dispersive X-ray spectroscopy (EDX), respectively.

Synthesis of Gd 2 O 3 Nanoparticles
The Gd 2 O 3 nanoparticles were synthesized using the following hydrothermal procedure. First, 0.5 mL of C 2 H 5 NH 2 was added slowly to a beaker containing an aqueous solution of 0.08 M Gd(NO 3 ) 3 ·6H 2 O under a vigorous stirring solution. The white precipitate solution obtained was heated at 120 • C for 12 h in an oven. The resulting mixture solution was centrifuged several times and then washed with deionized water to obtain gadolinium hydroxide, which was dried at 80 • C in an oven. Gd(OH) 3 was annealed in an electric furnace at 700 • C for 4 h in an Ar atmosphere to obtain Gd 2 O 3 [24].
3.2.2. Photocatalytic Activity of Gd 2 O 3 Nanoparticles for Degradation of Azo Dyes Such as MO, AO7, and AY23 Stock dye solutions were prepared using 2 mM of the azo dye powders. Next, 0.5 g/L of Gd 2 O 3 was used for the photocatalytic degradation of each 0.042-0.126 mM azo dye aqueous solution containing H 2 O 2 . The pH of the dye solutions was fixed at 5.5.
The solution was stirred with a magnetic bar for 30 min in the absence of light to attain an adsorption-desorption equilibrium between the dye molecules and the catalyst.
The photocatalytic degradation process was monitored by using UV-vis spectroscopy. Equation (3) was used to calculate the percentage of photocatalytic degradation of azo dyes. Additionally, the reaction kinetics and factors affecting the photocatalytic degradation of azo dyes were examined.

Photocatalytic Activity of Degradation of Azo Dyes Uner UV Irradiation at 254 nm and 365 nm
The photocatalytic degradation of MO, AO7, and AY23 by Gd 2 O 3 nanoparticles was carried out in aqueous dye solutions under UV irradiation at 254 nm and 365 nm. The initial concentration of azo dye solutions was 0.042 mM and the initial H 2 O 2 concentration was 4.8 M. The Gd 2 O 3 nanoparticles were added to the azo dye solutions, which were irradiated with UV light (output power: 4 W-254 nm, 4 W-365 nm) for 10, 20, 30, and 100 min.

Reusability of Gd 2 O 3 Nanoperticles as Photocatalyst for Degradation of MO
The photocatalytic degradation of MO with Gd 2 O 3 nanoparticles was performed by repeating the experiments five times. The photocatalytic experiment was carried out under the following conditions: a 0.042 mM initial MO concentration; a 4.8 M H 2 O 2 concentration; a 1.0 g/L photocatalyst dose; and a 100 min reaction time. After the photocatalytic degradation of MO using the Gd 2 O 3 nanoparticles as a catalyst, the photocatalyst of the Gd 2 O 3 nanoparticles was separated from the reaction mixture by centrifugation, washed with distilled water, and dried in an oven. Experiments on recycling were also carried out under the same conditions. To compensate for the catalyst loss during the washing process, a constant amount of catalyst concentration was maintained in each cycle test, and numerous degradation tests were conducted simultaneously at each recycle test, resulting in adequate catalysts being collected.

Conclusions
Under UV light irradiation, photocatalytic degradation of azo dyes such as MO, AO7, and AY23 was conducted in aqueous solutions with the Gd 2 O 3 nanoparticles and H 2 O 2 . The generation of ·OH and ·O 2 radicals, which can oxidize the azo dye molecules, causes photocatalytic degradation of the azo dyes on the catalyst surface. The degradation rate of the azo dyes increased when the initial dye concentration decreased and when the catalyst dose increased. For the degradation of azo dyes, the photocatalytic effectiveness of the Gd 2 O 3 nanoparticles declined in the following order: MO > AO7 > AY23. The photocatalytic degradation of the azo dye solutions under UV irradiation at 365 nm was higher than that under UV irradiation at 254 nm and was found to follow the pseudo first-order reaction rate law. The photocatalytic degradation of azo dye aqueous solutions was increased by increasing the UV irradiation time. More hydroxyl radicals are generated when Gd 2 O 3 /H 2 O 2 /UV light is combined, which improves the photocatalytic degradation of azo dyes. The reusability results demonstrate a small decrease in photocatalytic degra-dation of the azo dye solution for each cycle, with no apparent decrease in photocatalytic degradation after five cycles, demonstrating the stability of the Gd 2 O 3 nanoparticles.