*2.1. Characterization of Nb2O5 Photocatalyst*

The synthesized Nb2O5 photocatalyst powder was characterized by SEM. The micrograph in Figure 1a exhibits a uniform particle size distribution of ca. 200 nm. The magnified catalyst surface depicted in Figure 1b exhibits a high roughness and porosity, which could be induced by the calcination process of the niobium oxalate complex ammonium salt. The mapping of the chemical composition of the catalyst particles (Figure 1c) indicated that they are composed of niobium with a homogenous distribution.

**Figure 1.** Scanning electron micrographs of Nb2O5-synthesized photocatalyst with magnifications of (**a**) 500× and (**b**) 1000×, and (**c**) niobium mapping micrograph.

The FTIR spectrum of Nb2O5 in Figure 2 depicts the characteristic bands of niobium oxides: one shoulder at 900 cm−<sup>1</sup> is attributed to the stretch Nb-O and the bands at 661 and 600 cm−<sup>1</sup> are attributed to the angular vibration Nb-O-Nb [27]. Additionally, a band can be observed at 1622 cm−<sup>1</sup> that is related to the water adsorbed on the surface of Nb2O5 [8,28], and a small band can be observed at 1531 cm−<sup>1</sup> that is usually associated with impurities from the precursor salt of niobium. These impurities in the precursor are considered to be beneficial in the literature, since they act as stabilizers of the niobium oxide catalyst [29,30].

**Figure 2.** FTIR spectra of synthesized Nb2O5.

Niobium oxide has a complex crystal morphology, wherein at least 12 crystallographic structures have been identified to date [31]. The X-ray diffraction pattern obtained from the synthesized Nb2O5 powder catalyst is shown in Figure 3a. It can be seen that the diffractogram presents reflections 2θ = 22.6◦, 28.5◦, 36.7◦, 46.3◦, 50.4◦, and 55.3◦, which correspond to the crystallographic planes of miller index (001), (100), (101), (002), (110), and (102), respectively. These planes are characteristic of the pseudohexagonal structure of niobium oxide [27,32], proving the obtention of a catalyst with

a defined crystalline microstructure with a crystallite size of 2.2 nm estimated from the Scherrer formula. The electronic spectroscopy of diffuse reflectance made it possible to determine the band-gap energy required for the photo-promotion of one electron (see Figure 3b). The results indicated that Nb2O5 presents a band gap of 3.1 eV, similar to the characteristic band gap of titanium dioxide photocatalysts [9,33], and at the low end of the Nb2O5 bandgap values [34].

**Figure 3.** (**a**) X-ray diffractogram of Nb2O5. (**b**) UV-vis DRS absorption spectra of synthesized Nb2O5. The inset panel shows the Tauc plot for the band-gap energy determination of 3.13 eV.

The isotherm of nitrogen adsorption-desorption of Nb2O5 in Figure 4a presents a hysteresis loop characteristic of type IV isotherms, which is typical of mesoporous materials. This is in agreement with the characteristic morphologies observed by SEM in Figure 1. The specific surface area of 42.36 m2 g−<sup>1</sup> was calculated from the BET method (SBET). Meanwhile, the BJH analysis (Figure 4b) revealed the presence of uniformly sized mesopores with an average diameter (dp) of 10.1 nm and a mean pore volume (Vp) equal to 0.102 cm<sup>3</sup> g−1. These results are in agreement with those reported by Wang et al. [32] for commercial niobium oxide calcined at a temperature of 500 ◦C (SBET = 49.9 m<sup>2</sup> g−1, Vp = 0.12 cm<sup>3</sup> g−<sup>1</sup> e dp = 9.6 nm), even when the niobium oxide precursor selected was different.

**Figure 4.** (**a**) Nb2O5 photocatalyst nitrogen (-) adsorption–(-) desorption isotherms. (**b**) Pore size distribution plot that depicts the characteristic response of mesoporous materials.

#### *2.2. Photocatalytic Activity on Azo Dye Decolorization*

The photocatalytic activity of the synthesized Nb2O5 was verified on the basis of the corresponding decolorization of solutions containing 5 mg L−<sup>1</sup> methyl orange (MO) as model azo dye (see Table 1). MO is highly photostable, and is not degraded under direct sunlight irradiation, as depicted in Figure 5 [12]. Meanwhile, ca. 6.0% decolorization after 100 min was observed when 1.0 g L−<sup>1</sup> of Nb2O5 catalyst suspended in solution was exposed to sunlight irradiation. Photocatalytic degradation

can be assumed, since only a discrete 0.6% removal was observed under dark conditions due to adsorption on the porous Nb2O5. The slight photocatalytic removal under solar irradiation could be justified by the faster recombination Reaction (3), which diminishes the available oxidants (i.e., hvb<sup>+</sup> and -OH) [17,22]. The strategic use of selective ecb<sup>−</sup> scavengers may synergistically enhance the photocatalytic response by inhibiting the extent of Reaction (3) [16,26]. As shown in Figure 5, the addition of H2O2 boosts the performance of Nb2O5, which attains complete decolorization after 40 min. It is important to remark that the sole addition of H2O2 under dark conditions had no effect on dye solution decolorization, because of the weak oxidative capacity of H2O2 (Eº(H2O2/H2O) = 1.76 V/SHE). Nevertheless, decolorization was observed in the presence of H2O2 under direct solar irradiation due to the photolytic decomposition of H2O2 according to Reaction (7). Please note that this reaction only occurs under UVC irradiation, which is a component of the solar light in Northeast Brazil [24,35].

$$\text{H}\_2\text{O}\_2 + h\nu \to 2\,\,\text{^{\bullet}OH} \tag{7}$$

**Property Characteristics** IUPAC name Sodium 4-{[4-(dimethylamino) phenyl] diazenyl} benzene-1-sulfonate Common name Methyl Orange CAS number 547-58-0 Color Index number 13,025 M/g mol−<sup>1</sup> 327.3 λmax/nm 464 pKa 3.45 Chemical formula C14H14N3SO3Na Chemical structure **1 1 1 62**1D 

**Table 1.** Chemical structure and characteristics of Methyl Orange azo dye.

**Figure 5.** Evaluation of the photocatalytic degradation of 100 mL of 5 mg L−<sup>1</sup> of MO at pH 5.0 with (-, -) 1.0 g L−<sup>1</sup> of Nb2O5, (, -) 0.20 M H2O2, ( , ) Nb2O5 g L−<sup>1</sup> with 0.20 M H2O2, (-, , ) in dark conditions or (-, -, ) under sunlight irradiation. The photostability of the dye solution was also tested (+).

The synergetic effect observed between Nb2O5 and ecb<sup>−</sup> scavenger H2O2 can be explained by the enhanced generation of -OH from Reaction (2) due to the significant reduction of the recombination rate of Reaction (3) [26,36]. This effect results in a consequent increase in the availability of reactive oxygen species on the Nb2O5 photocatalyst surface, and then the increased oxidation capabilities of the system degrading the azo dye molecule [17,23].

The synergetic effect of H2O2 evidences its role in Nb2O5 photocatalytic efficiency performance. For this reason, the role of H2O2 concentration in solution was studied as a rate driving parameter for decolorization. Figure 6 shows the percentage of color removal attained for increasing doses of H2O2 acting as ecb<sup>−</sup> scavenger. Higher color removal percentages of 6.0%, 71.5% and 91.0% were observed for increasing concentrations of H2O2 of 0 M, 0.10 M and 0.20 M, respectively. It should be noted that further increase in H2O2 concentration resulted in a decrease in performance, achieving a lower color removal of only 77.2% after 80 min of Nb2O5 photocatalytic treatment. This phenomenon can be explained by the acceleration of concomitant waste reactions [17,37]. One of the main processes that decreases performance is the oxidation of excess H2O2 by -OH following Reaction (8) [24,26]. This undesired side-reaction not only consumes -OH, which will subsequently not be able to oxidize the target azo dye, but it also diminishes the available amount of H2O2. Moreover, the excessive accumulation of radical species can promote their dimerization following Reactions (9) and (10) [12,38]. These undesired reactions do not contribute to the overall photocatalytic performance, and may slow down the decolorization kinetics, as was observed experimentally (see Figure 6). Therefore, an optimal dose of 0.2 M H2O2 was defined for the following experiments to ensure faster solution decolorization and higher photocatalytic efficiency.

$$\text{H}\_2\text{O}\_2 + ^\bullet \text{OH} \rightarrow \text{HO}\_2\text{}^\bullet + \text{H}\_2\text{O} \tag{8}$$

$$\rm{HO}\_2\rm{}^\bullet + \rm{^\bullet OH} \rightarrow \rm{H}\_2\rm{O} + \rm{O}\_2\tag{9}$$

$$2\,\mathrm{^{\bullet}OH} \to \mathrm{H\_2O\_2} \tag{10}$$

**Figure 6.** Impact of the ecb<sup>−</sup> scavenger dose on the solar photocatalytic decolorization of 100 mL of 5 mg L−<sup>1</sup> of MO with 1.0 g L−<sup>1</sup> of Nb2O5 at pH 5.0. Initial H2O2 concentration: (-) 0 M, (-) 0.10 M, () 0.20 M, and () 0.30 M.

#### *2.3. E*ff*ect of pH on the Photocatalytic Decolorization of Azo Dye Methyl Orange*

The pH of the treated solution is one of the variables with the greatest influence on the photocatalytic degradation of pollutants, because it affects several physical-chemical properties of the catalysts that enhance or reduce the degradation efficiency, including the catalyst surface charge and the organic adsorptivity [17,23,39]. The point of zero charge (PZC) pHPZC = 4.86 of Nb2O5 was determined by the pH drift method [24,40], as described in the methodology section. The pHPZC is an intrinsic

characteristic of the catalyst that makes it possible to determine whether the surface of Nb2O5 is negatively or positively charged as a function of the pH. If the working pH conditions are above of the pHPZC of the Nb2O5, the catalyst surface is negatively charged. Conversely, the surface will be positively charged when the pH is below the pHPZC.

Figure 7 shows the effect of the initial pH on the decolorization efficiency of MO using 1.0 g L−<sup>1</sup> of Nb2O5 photocatalyst and 0.20 M of H2O2. It is important to note that in the range of pH under consideration, the sulfonic group of azo dye MO is deprotonated, with the pollutant molecule being negatively charged (see Table 1). However, at pH = 3.0, one of the molecule's N is protonated, and the global charge of the molecule will be neutral (pKa = 3.45). This consideration is an important fact that could justify the lower decolorization achieved at alkaline pH. Above the pHPZC, the Nb2O5 surface is negatively charged, and the dye molecule will also be negatively charged, thus leading to electrostatic charge repulsion, making the adsorption processes more difficult, along with the approach of the molecule towards the photocatalyst surface. Thus, with increasing values of pH, the number of negatively charged sites on the Nb2O5 also increases, further reducing the photodecolorization process in the following sequence 7.0 > 9.0 > 11.0, with percentages of color removal of 39.3%, 16.0% and 7.5%, respectively. In addition, it is well known that the rate of decomposition of unstable H2O2 increases with increasing pH in accordance with Equations (11) or (12) in strongly alkaline media [41]. This loss of H2O2 by chemical decomposition affects the overall photocatalytic performance. First, it reduces the availability of ecb- scavenger in solution. Second, it consequently reduces the generation of -OH radicals. In contrast, at pH 3.0, the surface is positively charged. This stimulates the molecules' approach to and absorption onto the surface through the negatively charged sulfonic group. Thus, below pH 3.0, complete color removal can be achieved at lower treatment times of 10 min.

$$2\ \text{H}\_2\text{O}\_2 \to 2\ \text{H}\_2\text{O} + \text{O}\_2\tag{11}$$

$$\mathrm{HO}\_{2}^{-} \to \mathrm{OH}^{-} + \frac{1}{2}\,\mathrm{O}\_{2} \tag{12}$$

**Figure 7.** Influence of the initial pH on the percentage of color removal during solar photocatalytic treatment of 100 mL of 5 mg L−<sup>1</sup> of MO with 1.0 g L−<sup>1</sup> of Nb2O5 0.2 M of H2O2 at pH: () 3.0, (-) 5.0, () 7.0, (-) 9.0 and () 11.0.

The nearest working pH to pHPZC = 4.86 was 5.0. Under these conditions, the catalyst surface is practically neutral and has no electro-attractive or electro-repulsive effects that affect the adsorption processes. Thus, as observed in Figure 7, it presents faster decolorization kinetics than higher, more alkaline pH, but slightly slower kinetics than those obtained for pH 3.0. The pH 5.0 was considered the optimum condition, because it is the natural pH of MO solutions and the nearest to the circumneutral pH of water effluents. This would decrease the potential environmental health and safety risks, as well as the costs to small/mid-sized industry, by minimizing the handling and storage of acids and bases. Furthermore, this may have an impact on operational costs, since it would obviate acidification for the treatment and neutralization steps prior to the release of treated effluent.
