*2.3. Influence of Photocatalyst Concentration on OII Removal*

From the previous results, ZnO–Ag with 1.3% (*w*/*w*) of Ag was selected as the most effective photocatalyst. The effect of ZnO–Ag (1.3%) concentration was evaluated in 10 mL of aqueous samples containing an initial OII concentration (COII,i) of 10 mg L−1, which were subjected to white light with 200–1000 mg L−<sup>1</sup> of ZnO NPs or ZnO–Ag (1.3%) NCs, and to UVA light with 50–500 mg L−<sup>1</sup> of photocatalyst (Figure 5). After 3 h of white light irradiation, samples containing from 200 to 750 mg L−<sup>1</sup> of ZnO–Ag (1.3%) showed increasing photocatalytic rates, obtaining 49 to 78% of OII elimination, respectively. This can be explained by the enhancement of the active sites in the catalyst by increasing their concentration in the samples.

**Figure 5.** Influence of photocatalyst concentration under white light irradiation for 180 min (**left**) and UVA light irradiation for 60 min (**right**). Error bars were calculated considering a normal distribution, for *p* < 0.01 [obtained from kinetic data].

After 5 h of white light irradiation of the sample containing ZnO–Ag (1.3%) NCs at 750 mg L<sup>−</sup>1, it was found that the OII concentration was below the detection limit of the spectrophotometric method, which was evidenced by the complete color removal of the sample. However, by increasing the concentration of the photocatalyst to a value of 1000 mg L−<sup>1</sup> of ZnO–Ag (1.3%), the extent of decolorization was inferior to that of the maximum: 60% OII removal. This phenomenon was observed by other authors for different photocatalysts [27,28] and is probably due to the effects of the NP aggregation and the reduction of the available surface area for photon absorption. A similar trend, but with slower reaction kinetics, was observed in samples containing ZnO NPs. As expected, the tests carried out under UVA light showed full removal of OII after 1 h, using ZnO–Ag (1.3%) NCs at 200 mg L<sup>−</sup>1. These data were adjusted following a second order equation with an adequate fitting of data, achieving R<sup>2</sup> up to 0.92.

Accordingly, the negative natural logarithm of the ratio between OII concentration and its initial concentration, ln (COII,0/COII,t), was plotted as a function of the irradiation time and a linear regression was obtained. Correlation coefficients (R2), half-lives (t1/2) and apparent pseudo-first order rate constants (k) are presented in Table 1. R<sup>2</sup> ranges from 0.9323 to 0.9990, confirming the suitability of the pseudo-first order model to describe the kinetics of OII removal in the presence of ZnO–Ag (1.3%) and ZnO NPs, also applied by other authors to model the photocatalytic degradation of dyes and emerging contaminants. [16,27,29–31].


**Table 1.** Kinetic parameters for the photodegradation of OII with ZnO–Ag (1.3%) and ZnO NPs.

<sup>a</sup> Not significantly different (*p* < 0.05).

Other works reported the photocatalytic degradation of OII using modified ZnO catalyst to improve kinetics. Chen et al. [27] obtained a kinetic constant of 0.033 h−<sup>1</sup> using micro-structured ZnO, which amounted to 0.065 h−<sup>1</sup> for ZnO decorated with Ag, both values lower than those obtained in these experiments, and they also used a larger catalyst concentration: 1500 mg L<sup>−</sup>1. The enhancement of the results using nanostructured ZnO could take place because nanostructured catalysts have more surface/volume ratio than the micro-sized material, increasing the number of active sites per mass unit. Moreover, Siuleiman et al. [30] have structured ZnO in nanowires, obtaining kinetic constants of 0.092 and 0.112 h−<sup>1</sup> for UVA and visible light irradiation with a catalyst concentration of 500 mg L<sup>−</sup>1, respectively.

In view of the above, the incorporation of the Ag nanoclusters causes an improvement of the kinetic constants of 3–6 times, both under UVA and white light radiation. In addition, the improvement in degradation rates by comparing the same catalyst concentrations under white and UVA light is 7–10 times greater. For all cases, the most notable differences occur for a catalyst concentration in the range of 200–500 mg L−1. These ratios are similar to those obtained by Sornalingam et al. [31] using Au-TiO2 NCs with UVA and cold white light. Reuse tests cannot be carried out due to the losses of catalyst at the recovery stage.
