Optimization of Solar Photocatalytic Degradation of Chloroxylenol Using TiO₂, Er³⁺/TiO₂, and Ni²⁺/TiO₂ via the Taguchi Orthogonal Array Technique

Abstract

In this study, the optical properties of a TiO₂ photocatalyst were enhanced with various impregnations of Er³⁺ and Ni²⁺ separately, using the impregnation method as photocatalysts for the direct solar photolysis degradation of chloroxylenol. The synthesized Er³⁺/TiO₂ and Ni²⁺/TiO₂ catalysts were characterized using X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM), energy dispersive X-ray (EDX), metal mapping, and ultraviolet visible (UV–Vis) spectroscopy. The results showed that the Er³⁺/TiO₂ and Ni²⁺/TiO₂ nano-particles have the same structures of TiO₂ nano-particles with little difference in particle size. The Er³⁺ and Ni²⁺ ions were well-distributed on the TiO₂ surface, and it was found that the maximum band gap decreased from 3.13 eV for intrinsic TiO₂ to 2.63 eV at 1.8 wt % Er³⁺/TiO₂ and to 2.47 eV at 0.6 wt % for Ni²⁺/TiO₂. The initial concentration of chloroxylenol, catalyst loading, and pH of the solution are the most important factors affecting the solar photocatalytic degradation efficiency that were optimized using Design Expert software (version 6.0.10, Minneapolis, MN, USA, 2003). The results showed that the optimal conditions for chloroxylenol degradation include a pH of 4, TiO₂ loading at 3 g/L, and a chloroxylenol concentration of 50 mg/L. These conditions resulted in a degradation efficiency of 90.40% after 60 min of direct solar irradiation, wherein the solar energy recorded during a clear sunny day is 1000 W/m². However, some experiments were conducted on a semi-cloudy day to cover all weather stated and to study the degradation kinetics. During semi-cloudy day experiments, using Er³⁺/TiO₂ and Ni²⁺/TiO₂ with a solar irradiation activity of 600 W/m² for a 60 min exposure at optimal conditions increased the degradation efficiency from 68.28% for intrinsic TiO₂ to 82.38% for Er³⁺/TiO₂ and 80.70% for Ni²⁺/TiO₂.

1. Introduction

One of the most pervasive problems that people face today is the effluent of toxic compounds in water sources. One of these compounds is chloroxylenol (4-chloro-3,5-dimethylphenol), which is widely used as an antibacterial agent. It is a phenolic derivative used in many product formulations for use as an antiseptic and disinfectant [1]. Chloroxylenol has unique antiseptic properties and is a very effective antimicrobial agent against many germs that cause infections since it disrupts the proton gradient of the cell membrane necessary for the bacteria to produce adenosine triphosphate (ATP). The ATP deficiency results in cell death due to starvation [2]. Chloroxylenol also oxidizes the structure of the cell and impedes the nutrients from passing through the cell wall, causing a loss of normal enzyme activity leading to cell death. Furthermore, chloroxylenol also changes the permeability of the cell walls of microorganisms and disrupts its biological processes [3]. Owing to its antimicrobial properties, chloroxylenol is widely used in various applications and is present in many products and formulations. Due to the large use of products that contain chloroxylenol in their formulations, it can be discharged directly or via urban sewage systems into aquatic systems. Chloroxylenol, as a biocide, represents a potential hazard and has adverse impacts on human health and the environment [4,5]. Due to this impact, efforts are focused on the removal of this compound from aqueous solutions by coagulation [6], adsorption [7], oxidation [8], and biodegradation [9]. All of these methods have significant limitations and disadvantages. Solar photocatalytic oxidation is a promising method because of the low cost of the energy and most of the pollutants can be completely mineralized to CO₂ with suitable catalysts. TiO₂ is one of the best photocatalysts because it is highly photosensitive, non-toxic, has strong oxidizing power and long-term stability. One of the disadvantages of TiO₂ nano-particles (NPs) is their high band gap energy (3.2 eV). Hence, the doping of TiO₂ NPs with metal (rare earth) ions, wherein these metal ions will become concentrated on the surface of the TiO₂ NPs, is one way to decrease the band gap energy and enhance the photoactivity of the TiO₂ NPs [10]. The absorption of visible light by TiO₂ becomes enhanced because visible light makes up approximately 43% of the solar spectrum as its primary source of energy [11]. However, many researchers have doped TiO₂ with rare earth (RE) and transition metal ions for specific applications in the ultraviolet (UV) and visible frequency ranges [12,13]. A design of experiment (DOE) is an effective method to reduce the total cost of the process by minimizing the number of experimental runs. Many designs, such as the Box–Behnken, Taguchi, D-optimal, and central composite designs, have been used to improve the efficiency of degradation [14]. Recently, the Taguchi method has been reported as the ideal method for the optimization process, which is a simple and easy tool that provides effective solutions [15]. Fewer experimental runs with a higher performance value of the response close to the target value have been obtained by the Taguchi method compared with other methods [16]. To optimize any process using the optimization design, it is necessary to identify all factors that have the highest effect on an existing process. When the Taguchi method is applied, the following two points have to occur: (1) clearly define a set of orthogonal arrays (OA), each of which can be used for many experimental situations; and (2) devise a standard method for the analysis of the results [17,18,19]. Optimization via the Taguchi method can significantly reduce the time required for experiments, which is used to show the effects of multiple factors on performance and determine which factor has more or less influence on the process [20]. In the Taguchi method, after the generation of the optimum conditions, the experimental results should be confirmed and compared with the prediction results [21]. As a continuation of previous studies [22,23,24,25,26,27,28] on nanomaterials and catalysts, herein, we report the solar photocatalytic degradation of chloroxylenol that was optimized by the DOE Taguchi OA using TiO₂ NPs, as well as Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs prepared by the impregnation method.

2. Results and Discussion

2.1. X-ray Diffraction (XRD) of TiO₂, Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs

All prepared samples were subjected to X-ray diffraction (XRD) to investigate the type of crystallinity of the used materials and to determine if any change that occurred during the impregnation of the TiO₂ NPs with Er³⁺ and Ni²⁺ ions. The XRD scanning of the Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs is, respectively presented in Figure 1 and Figure 2. The peaks that appear for both the anatase and rutile types (JCPDS File No. 21-1276) of TiO₂ [29,30] are clear. These patterns show that the structure and phase compositions of the TiO₂ NPs were not affected by impregnation with Er³⁺ and Ni²⁺ ions. Furthermore, the intensity of all TiO₂ peaks decreased, which means that the crystal structure of the TiO₂ NPs is impregnated with Er³⁺ and Ni²⁺ ions.

Figure 1. X-ray diffraction pattern: (a) TiO₂; (b) 0.6 wt % Er³⁺/TiO₂; (c) 1.2 wt % Er³⁺/TiO₂; (d) 1.8 wt % Er³⁺/TiO₂; and (e) 2.4 wt % Er³⁺/TiO₂.

Figure 2. X-ray diffraction pattern: (a) TiO₂; (b) 0.2 wt % Ni²⁺/TiO₂; (c) 0.4 wt % Ni²⁺/TiO₂; (d) 0.6 wt % Ni²⁺/TiO₂; and (e) 0.8 wt % Ni²⁺/TiO₂.

Table 1 and Table 2 show the values of the intensity for each of the peaks. Diffraction spectra of the Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs samples did not indicate the presence of Er or Ni. This result can most likely be attributed to the favorable dispersion of Er³⁺ and Ni²⁺ in the TiO₂ structural lattice [31].

Table 1. Intensity of all peaks of TiO₂ and Er/TiO₂.

2θ	Intensity (Arbitrary Unit, a.u.)
	TiO2	0.6 wt % Er/TiO2	1.2 wt % Er/TiO2	1.8 wt % Er/TiO2	2.4 wt % Er/TiO2
25.2712	2158	1984	1911	1879	1855
27.4136	515	498	493	481	474
37.3038	127	105	100	90	90
48.0908	555	536	481	478	473
53.8206	311	310	288	262	230
53.97	379	368	351	346	336
55.116	367	340	313	291	283
62.7142	292	285	272	249	241
68.8675	165	149	143	131	120
70.2128	145	114	109	108	104
75.0956	185	164	163	151	147

Table 2. Intensity of all peaks of TiO₂ and Ni/TiO₂.

2θ	Intensity (Arbitrary Unit, a.u.)
	TiO2	0.2 wt % Ni/TiO2	0.4 wt % Ni/TiO2	0.6 wt % Ni/TiO2	0.8 wt % Ni/TiO2
25.2712	2158	1964	1922	1866	1834
27.4136	515	504	464	440	410
37.3038	127	110	100	108	108
48.0908	555	536	514	509	483
53.8206	311	284	284	282	270
53.9700	379	348	347	343	316
55.1160	367	334	329	319	294
62.7142	292	254	254	253	242
68.8675	165	144	138	133	108
70.2128	145	122	113	111	106
75.0956	185	178	160	156	141

The catalyst crystallite size D was determined according to Debye–Scherer equation (Equation (1)):

$$\mathit{D}=\frac{K\mathsf{\lambda }}{\mathsf{\beta } \mathit{c}\mathit{o}\mathit{s}\mathsf{\theta }}$$

(1)

where K is the Scherer constant (K = 0.89), λ is the incident X-ray wavelength, β is the peak width at half maximum, and θ is the Bragg diffraction angle. The data of the catalyst crystallite size is presented in Table 3 for TiO₂ with several wt % of Er³⁺ and Ni²⁺. It is observable that there is a small increase in particle size as impregnation increases for Er³⁺/TiO₂ and Ni²⁺/TiO₂ samples.

Table 3. Catalyst crystallite size of TiO₂ with several wt % of Er³⁺ and Ni²⁺.

Sample	Crystallite (nm)
TiO2	28.72
0.6 wt % Er3+/TiO2	28.84
1.2 wt % Er3+/TiO2	31.53
1.8 wt % Er3+/TiO2	33.93
2.4 wt % Er3+/TiO2	35.93
0.2 wt % Ni2+/TiO2	28.98
0.4 wt % Ni2+/TiO2	30.84
0.6 wt % Ni2+/TiO2	31.10
0.8 wt % Ni2+/TiO2	31.51

2.2. Catalyst Surface Morphology of TiO₂, Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs

The morphologies of the Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs were investigated using field emission scanning electron microscopy (FESEM), energy dispersive X-ray (EDX), and metal mapping. The FESEM images are shown in Figure 3. The images revealed that the shape of the particles are mostly spherical, and less agglomeration was observed in the Er³⁺/TiO₂ sample than the Ni²⁺/TiO₂ sample.

Figure 3. Field emission scanning electron microscopy images: (a) TiO₂; (b) 0.6 wt % Er³⁺/TiO₂; (c) 1.2 wt % Er³⁺/TiO₂; (d) 1.8 wt % Er³⁺/TiO₂; (e) 2.4 wt % Er³⁺/TiO₂; (f) 0.2 wt % Ni²⁺/TiO₂; (g) 0.4 wt % Ni²⁺/TiO₂; (h) 0.6 wt % Ni²⁺/TiO₂; and (i) 0.8 wt % Ni²⁺/TiO₂.

EDX analysis was conducted on the Er³⁺/TiO₂ and Ni²⁺/TiO₂ samples to investigate the presence of Er and Ni on the TiO₂ surface. The results of the EDX spectra are shown in Figure 4 and show that the Er and Ni are present in each of their respective samples. Additionally, the EDX spectral data provided the wt % of the Er and Ni content in the samples, and the results are summarized in Table 4. These results indicate that there is good agreement between the calculated and actual content of Er and Ni. Metal mapping tests were conducted to investigate the dispersion of Er and Ni on the TiO₂ surface. The results are shown in Figure 5 and reveal a good distribution of Er and Ni on the TiO₂ NPs in each of the prepared samples.

Figure 4. Energy dispersive X-ray spectra: (a) 0.6 wt % Er³⁺/TiO₂; (b) 1.2 wt % Er³⁺/TiO₂; (c) 1.8 wt % Er³⁺/TiO₂; (d) 2.4 wt % Er³⁺/TiO₂; (e) 0.2 wt % Ni²⁺/TiO₂; (f) 0.4 wt % Ni²⁺/TiO₂; (g) 0.6 wt % Ni²⁺/TiO₂; and (h) 0.8 wt % Ni²⁺/TiO₂. Red circle: Er³⁺; green circle: Ni²⁺.

Table 4. Calculated and actual wt % of Er and Ni on TiO₂ surface using energy dispersive X-ray (EDX).

Sample	Calculated Metal (wt %)	Actual Metal (wt %) by EDX
0.6 wt % Er3+/TiO2	0.6	0.72
1.2 wt % Er3+/TiO2	1.2	1.31
1.8 wt % Er3+/TiO2	1.8	2.06
2.4 wt % Er3+/TiO2	2.4	2.45
0.2 wt % Ni2+/TiO2	0.2	0.16
0.4 wt % Ni2+/TiO2	0.4	0.31
0.6 wt % Ni2+/TiO2	0.6	0.59
0.8 wt % Ni2+/TiO2	0.8	0.74

Figure 5. Metal mapping (a) Er³⁺/TiO₂ and (b) Ni²⁺/TiO₂.

2.3. Absorbance and Band Gap of TiO₂, Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs

The absorbance spectra for all the prepared samples (0.6, 1.2, 1.8, and 2.4 wt %) and (0.2, 0.4, 0.6, and 0.8 wt %) of Er³⁺/TiO₂ and Ni²⁺/TiO₂, respectively, were recorded at room temperature and are shown in Figure 6. The maximum absorption was observed at 320 nm. The direct band gap (E_g) values of TiO₂, Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs were determined using the following direct transition equation (Equation (2)):

$$\mathsf{\alpha }\mathrm{h}v={\mathrm{E}}_{\mathrm{d}}{\left(\mathrm{h}v-E_{\mathrm{g}}\right)}^{\frac{1}{2}}.$$

(2)

where (α) is the optical absorption coefficient, h is Planck’s constant, v is the frequency (hv = 1240/wavelength), E_g is the direct band gap, and E_d is a constant [32]. By plotting (αhv)² as a function of the photon energy (hv) and extrapolating the linear portion of the curve to a value of zero, band gap (E_g) was calculated and is shown in Figure 7. The obtained values of the band gap (E_g) of TiO₂ with several wt % of Er³⁺ and Ni²⁺ are listed in Table 5. The minimum values of the band gap energies obtained are 2.63 and 2.47 eV for Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs, respectively.

Figure 6. Ultraviolet visible spectra: (a) Er³⁺/TiO₂ nano-particles (NPs) and (b) Ni²⁺/TiO₂ NPs.

Figure 7. Direct band gap energy (E_g) calculation.

Table 5. Band gap (E_g) of TiO₂ with several wt % of Er³⁺ and Ni²⁺

Sample	Band Gap Energy (eV)
TiO2	3.13 ± 0.02
0.6 wt % Er3+/TiO2	3.20 ± 0.02
1.2 wt % Er3+/TiO2	3.00 ± 0.02
1.8 wt % Er3+/TiO2	2.63 ± 0.02
2.4 wt % Er3+/TiO2	2.70 ± 0.02
0.2 wt % Ni2+/TiO2	2.60 ± 0.02
0.4 wt % Ni2+/TiO2	2.50 ± 0.02
0.6 wt % Ni2+/TiO2	2.47 ± 0.02
0.8 wt % Ni2+/TiO2	2.61 ± 0.02

2.4. Experimental Design

2.4.1. Response of Solar Photocatalytic Degradation Experiments and Modeling

All experiments were conducted according to the Taguchi OA plan shown in Table 6 under constant direct solar radiation on a clear sunny day. All samples were exposed to solar radiation for a period of 60 min. The degradation efficiency (DE%) was calculated using Equation (1). The results of the DE% for all experiments are listed in Table 6. The results of the predicted values of the DE% generated by the model created by Design Expert software are in agreement with the experimental results presented in Table 6.

Table 6. Actual and predicted values of photocatalytic degradation efficiency percentage (DE%) of chloroxylenol.

Standard	Run Number	A: Catalyst Loading (g/L)	B: pH	C: Concentartion (mg/L)	DE% (Actual)	DE% (Predicted)
1	8	1	4	50	86.75	84.48
2	9	1	6	100	50.12	49.99
3	3	1	8	150	42.9	45.29
4	2	2	4	100	54.55	56.94
5	6	2	6	150	49.65	47.38
6	5	2	8	50	87.69	87.56
7	4	3	4	150	52.31	52.18
8	7	3	6	50	85.1	87.49
9	1	3	8	100	60.13	57.86

After the experimental results of the DE% analysis were obtained, the predicted model of the process was fit with a linear regression, and several factors were determined to have a significant effect on the estimated final model. The linear model of the process and the factors affecting the results and respective coefficients are shown in Equation (3) below:

DE% = +63.25 − 3.32 × A [1] + 0.72 × A [2] + 1.29 × B [1] − 1.62 × B [2] + 23.27 × C [1] − 8.31 × C [2]

(3)

where DE% represents the degradation efficiency determined by UV–Vis spectroscopy; A, B, and C are the catalyst loading, pH, and initial concentration, respectively.

2.4.2. Analysis of Variance (ANOVA)

The ANOVA results are summarized in Table 7. It is clear that the model is significant with a 95% confidence interval since the p-value was 0.0371, which is less than 0.05. The analysis revealed that the most significant factor is the initial concentration of the chloroxylenol. The ${\mathrm{R}}^2$ and ${\mathrm{R}}_{\mathrm{adj}}^2$ values, which were 0.9875 and 0.9499, respectively, show that a sufficient model was created to explain the optimization conditions for the degradation process. The adequacy precision was 11.87, which is larger than 4, indicating that the model is adequate.

Table 7. ANOVA analysis results for photocatalytic degradation of chloroxylenol ¹.

Source	Sum of Squares	Degree of Freedom	Mean Square	F-Value	p-Value
Model	2570.93	6	428.49	26.25	0.0371
A	54.95	2	27.48	1.68	0.3722
B	13.22	2	6.61	0.40	0.7120
C	2502.75	2	1251.38	76.66	0.0129
Residual	32.65	2	16.32	–	–

¹ Significant at p-value < 0.05, ${\mathrm{R}}^2$ = 0.9875, ${\mathrm{R}}_{\mathrm{adj}}^2$ = 0.9499, Standard Deviation = 4.04, mean = 63.25, Coefficient of Variation = 6.39, press = 660.50, adequacy precision = 11.87. ANOVA, analysis of variance.

The diagnostic plots of the optimization process are shown in Figure 8. The experimental data shows a good fit with the graphical line, confirming that the model provides a good assumption based on the ANOVA results as shown by the normal probability versus the Studentized residual plots in Figure 8a. The random scattering of the experimental points compared to the funnel-shaped pattern that was obtained indicates that the DE% response had an original observation of variance with no problem between the response and the factors, as shown in Figure 8b [33]. Figure 8c shows a good distribution of the experimental points; none were located out of the indicated range. The actual value of the DE% with the predicted value of the model was evaluated by the value of ${\mathrm{R}}^2$ and ${\mathrm{R}}_{\mathrm{adj}}^2$, as plotted in Figure 8d. All of the points in the figure are close to the line, indicating the capability of the model to predict the optimum conditions for the degradation process. The Studentized residuals for the experimental runs are shown in Figure 8e, which indicates that the model has values less than ±3.5 for the Studentized residuals, providing a good fit of the model to the response.

Figure 8. Diagnostic plots of the photocatalytic degradation efficiency percentage (DE%) for the optimization process. (a) Studentized residual; (b) Residual vs. predicted; (c) Outlier; (d) Actual vs. predicted; and (e) Residual vs. Run.

2.4.3. Optimal Conditions for Maximum DE% and Model Validation

To obtain the optimal conditions of the maximum degradation efficiency, the data shown in Table 6 was optimized numerically using Design Expert software for DE% (larger is better) for the ranges of all factors. The optimal conditions were found to be 3 g/L of Catalyst Loading, pH at 4 and 50 mg/L of chloroxylenol Concentration which the highest DE% of 90.40.

To validate the model shown in Equation (3) created by the software and the optimal conditions for the maximum DE%, three experimental runs were conducted under the optimal conditions listed in Table 8. The results of these experimental runs are shown in Table 8.

Table 8. Results of experimental runs for model validation.

Number of Experiment	Catalyst Loading (g/L)	pH	Concentration (mg/L)	DE%
1	3	4	50	89.57
2	3	4	50	88.21
3	3	4	50	87.67

The average DE% according to the experimental data shown in Table 8 is 88.48%, which is only a 1.92% difference from that of the value predicted by the model (90.40%) and, thus, supports the model validity.

2.5. Kinetics of the Photocatalytic Degradation of Chloroxylenol Using TiO₂ NPs

The experimental results of the photocatalytic degradation of chloroxylenol are shown in Figure 9. The reported results of the photocatalytic degradation of chloroxylenol show that the degradation rates of the photocatalytic oxidation using TiO₂ NPs could be fit with the first-order kinetic model [19]. Figure 10 shows the plot of ln(C₀/C) vs. the irradiation time for the degradation of chloroxylenol. The linearity of the plot suggests that the photodegradation reaction approximately follows the pseudo-first order kinetics with K = 0.0287 min⁻¹.

Figure 9. Photocatalytic degradation of chloroxylenol using TiO₂ NPs under optimal conditions.

Figure 10. Plot of the pseudo-first order rate constant of photocatalytic degradation of chloroxylenol using TiO₂ NPs under optimal conditions.

Application of Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs in the Photocatalytic Degradation of Chloroxylenol

To investigate the effects of the impregnation of TiO₂ NPs with Er³⁺ and Ni²⁺ on the DE%, experiments were performed using the optimal conditions shown in Table 8 on a semi-cloudy day (solar radiation 600 W/m²) with a constant reaction time of 60 min using different wt % of Er³⁺/iO₂ and Ni²⁺/TiO₂ NPs. The results for the DE% using Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs are shown in Figure 11 and Figure 12, respectively. Figure 11 shows the maximum DE% of 82.38% obtained at a 1.8 wt % of Er³⁺/TiO₂ provides a 14.10% increase compared with that of the intrinsic TiO₂ NPs at 68.28%. The maximum DE% of 80.70% obtained at a 0.6 wt % of Ni²⁺/TiO₂ showed an increase of 12.42% compared with that intrinsic TiO₂ NPs, as shown in Figure 12.

Figure 11. DE% of chloroxylenol using different wt % of Er³⁺/TiO₂.

Figure 12. DE% of chloroxylenol using different wt % of Ni²⁺/TiO₂.

In general, the DE% increased with the increasing concentration of both Er³⁺ and Ni²⁺ relative to the intrinsic TiO₂. The DE% increased up to an optimal concentration for both Er³⁺ and Ni²⁺ over intrinsic TiO₂ at 1.8 and 0.60 wt %, respectively. This result is attributed to the decrease in the band gap because of the shifting absorbance to the visible region. The band gaps of Er³⁺ and Ni²⁺, compared with intrinsic TiO₂, at 1.8 and 0.60 wt % were found to be 2.63 and 2.47 eV, respectively. These results were compared to intrinsic TiO₂ at 3.13 eV in Table 5. This shifting is attributed to the incorporation of Er³⁺ and Ni²⁺ into the surface of the TiO₂ nanoparticles [34]. After the optimal doping concentrations of Er³⁺ and Ni²⁺ for TiO₂, a decrease in the DE% was observed, which is attributed to the excess Er³⁺ and Ni²⁺ ions working as a recombination center within the degradation process [35]. However, the electron-hole recombination is faster in the presence of Er³⁺ and Ni²⁺ ions due to the unstable Er⁴⁺ and Ni³⁺ ions that can easily transfer electrons to the oxygen molecules [36].

3. Materials and Methods

3.1. Materials

All chemicals were purchased from Sigma Aldrich (St. Louis, MO, USA), including erbium (III) acetylacetonate hydrate, CAS grade, 97% purity; nickel (II) acetate tetrahydrate, 98% purity; 99.5% titanium oxide nano-powder (P25) with a diameter of 21 nm using transmission electron microscope TEM, (Philips, Eindhoven, The Netherlands); and chloroxylenol, 98% purity. All chemicals were used without further purification. Deionized water was used for the preparation of the chloroxylenol standard solution, as well as for dilutions.

3.2. Preparation of Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs

To impregnate the TiO₂ NPs with Er³⁺ and Ni²⁺ ions in different wt % of Er/TiO₂ (0.6, 1.2, 1.8, and 2.4 wt %) and Ni/TiO₂ (0.2, 0.4, 0.6, and 0.8 wt %), an 80-mL glass beaker was used. In total, 3 g TiO₂ powder was placed in the beaker. Particular amounts (60, 120, 180, and 240 mg) of erbium (III) acetylacetonate hydrate or nickel (II) acetate tetrahydrate were dissolved in 10 mL of deionized water and sonicated for 10 min. These solutions were then added to the TiO₂ powder dropwise under well-mixing conditions with a glass rod. The mixture was then stirred at 1300 rpm for 2 h, and the mixture was dried overnight at 80 °C, then calcined at 500 °C for 2 h.

3.3. Characterization of Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs

The crystallite size and phase composition of the Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs samples were determined using X-ray diffractometer (XRD) model D8 Advance Bruker AXS X-ray (Bruker, Karlsruhe, Germany) with Cu Kα radiation (1.5406 Å) in a 2θ scan range of 20°–80°.The morphology of the Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs was investigated using a field emission scanning electron microscope, model SUPRA 55 VP (Carl Zesis AG, Munich, Germany), under high resolution, supplied with the energy dispersive X-ray analysis (EDX) system and metal mapping model OXFORD INCA Penta FETx3 (Carl Zesis AG). The zeta potential of the Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs was measured using a Malvern Zeta sizer, model Nano-ZS (Malvern instruments Inc. Worcestershire, UK). A Perkin Elmer (lambda35) UV–Vis spectrophotometer (Perkin Elmer, Waltham, MA, USA) equipped with an integrating sphere was used to record the UV–Vis absorption spectra of the Er³⁺/TiO₂ and Ni²⁺/TiO₂ NPs. The spectra were recorded at room temperature with a wavelength range of 200–1100 nm.

3.4. Design of Experiment Using Taguchi Orthogonal Array L9 (3⁴)

The purpose of using the Taguchi Orthogonal Array design of experiment is to determine the optimum process conditions to obtain the maximum degradation efficiency (DE%) of chloroxylenol with the minimum number of experiments. In this study, Design Expert software version 6.0.10 was used. Three factors affecting the degradation efficiency of the chloroxylenol were investigated at three levels, as shown in Table 9.

Table 9. Factors and their levels.

Independent Variable	Factor	Level
		−1	0	+1
Catalyst loading	A	1	2	3
pH	B	4	6	8
Initial concentration	C	50	100	150

The orthogonal array L9 (3⁴) experimental design method was chosen to determine the experimental plan as given in Table 10.

Table 10. Experimental plan using the L9 (3⁴) orthogonal array.

Standard	Run Number	A: Catalyst Loading (g/L)	B: pH	C: Concentration (mg/L)
1	8	1	4	50
2	9	1	6	100
3	3	1	8	150
4	2	2	4	100
5	6	2	6	150
6	5	2	8	50
7	4	3	4	150
8	7	3	6	50
9	1	3	8	100

3.5. Photocatalytic Reactor and Experimental Procedures

A glass cylindrical slurry batch reactor with a 6.5-cm diameter, 3.8-cm height, 125-mL volume, and a 110-cm² radiation area was used to conduct the experiments of solar photocatalytic degradation of chloroxylenol. A 100-mL aqueous solution of a known concentration of chloroxylenol was mixed in the dark with a certain amount of TiO₂ NPs for 60 min to achieve adsorption-desorption equilibrium. The pH was adjusted using HCl and NaOH according to the experimental plan shown in Table 10. Most of the photocatalytic experiments were carried out under similar conditions on clear sunny days from 11:00 am to 2:00 pm with the ambient temperature between 28 and 40 °C, with the exception of those conducted to study kinetics that were run during semi-cloudy days. In each experiment, a sample was withdrawn every 15 min, centrifuged, and then filtered using a 0.22 μm filter to remove the catalyst particles prior to analysis.

3.6. Analytical Method

The UV–Vis spectrophotometer equipped with an integrating sphere was used to determine the concentration of chloroxylenol at λ = 297 nm. Five standard concentrations of chloroxylenol, 12.5, 25, 50, 100 and 150 mg/L, were used to determine the calibration curve. The calibration curve was defined by plotting the concentration versus its absorbance for the five standard concentrations. Then, the absorbance values of the chloroxylenol sample before and after degradation were recorded and compared with the calibration curve to determine their concentrations. The photocatalytic degradation efficiency (DE%) was calculated using the following Equation (4):

$$DE\%=\frac{\left(C_0-C\right)}{C_0} x$$

(4)

where C₀ is the initial concentration and C is the concentration of the chloroxylenol at different time intervals.

4. Conclusions

The TiO₂ nanoparticles were impregnated with various wt % of erbium and nickel ions using the impregnation method followed by calcination at 500 °C. The Er³⁺ and Ni²⁺ impregnated TiO₂ samples indicated that the band gap energy decreases as the wt % values of the Er³⁺ and Ni²⁺ increase up to 1.8 wt % for the Er³⁺/TiO₂ and 0.6 wt % for the Ni²⁺/TiO₂. Agglomeration and particle sizes increased as the ions’ wt % increased. The use of Design Expert software and the Taguchi OA method decreased the number of experiments required to predict the optimal conditions. The optimal conditions for solar photocatalytic degradation of chloroxylenol in an aqueous solution include a pH of 4, a TiO₂ loading of 3 g/L, and a concentration of 50 mg/L. The photocatalytic activity of TiO₂ showed obvious enhancements with only a small wt % increase upon impregnation with Er³⁺ and Ni²⁺ ions. This result was attributed to the decrease in the band gap energy. As a result, there is significant enhancement in the DE% of chloroxylenol by using the impregnated Er³⁺/TiO₂ and Ni²⁺/TiO₂.