Taguchi L₉ (3⁴) Orthogonal Array Design for Photocatalytic Degradation of Methylene Blue Dye by Green ZnO Particles Biosynthesized by Chrysanthemum spp. Flower Extract

Abstract

The pollution of synthetic dyes in wastewater exerts many negative impacts on the environment and human health. There is an increasing demand for the degradation of dyes, with an emphasis on photocatalysis. Here, we investigated the bio-mediated synthesis of ZnO using Chrysanthemum spp. flower extract and its utilization for the removal of methylene blue dye under sunlight irradiation. The bandgap energy of green ZnO nanoparticles was determined to be 3.0. The Taguchi L₉ (3⁴) orthogonal array design was applied to optimize the photocatalytic degradation of methylene blue dye by green ZnO particles. Four parameters, including the initial concentration (10–50 mg/L), ZnO dosage (0.33–1.0 mg), contact time (30–120 min), and pH (4–10) of the solution, were surveyed based on the Taguchi design. We found that the test result (99.0%) at 10 mg/L was almost equivalent to the predicted value (99.5%) of degradation efficiency. The reaction mechanisms shed light on the major role of reactive oxygen species (^•O₂⁻, ^•OH). More importantly, the green ZnO particles could be reused for at least five cycles and demonstrated high stability.

1. Introduction

The growth of the industrial sector has resulted in the discharge of large amounts of toxic waste into the environment, representing a globally critical problem [1]. For example, the synthetic dyes used for printing and textile and paper production are among the most heavily consumed organic compounds [2]. Over the past few decades, the considerable amount of synthetic dye in wastewater has caused many negative impacts on the environment and human health [3]. As a typical dye, methylene blue has shown toxic effects in humans, including carcinogenicity, hypertension, and skin irritation [4]. Some works have also reported that the presence of this dye in water significantly reduces the intensity of light, decreasing the photosynthesis performance of aquatic microorganisms [5]. It has been noted that methylene blue dye contributes to unbalancing aquatic ecosystems and lessening the quality of water sources [6]. As a result, the treatment of dye contaminants is worthy of consideration [7].

Zinc oxide (ZnO) nanoparticles possess many crystalline structures such as wurtzite, zinc blende, and rocksalt, with a wide range of bandgaps (3.1–3.8 eV) [8]. ZnO nanomaterials have demonstrated their promise in antibacterial, photocatalysis, adsorption, semiconductor, biosensor, nanofertilizer, and nanopesticide applications [9]. Many works have reported the excellent performance of ZnO particles in the photocatalytic degradation of pollutants [10]. For example, Ameen et al. [11] reported a high degradation efficiency of 93% against methylene blue dye using green ZnO synthesized from the Acremonium potronii fungal species. Using the hydrothermal method, Zaidi et al. [12] obtained ZnO nanospheres with a bandgap of 3.22 eV. The photocatalytic reaction of Bismarck brown dye reached a pseudo-first-order rate constant of 0.018 min⁻¹ and a degradation efficiency of 94%. Chen et al. [13] successfully synthesized ZnO photocatalysts by the sol–gel method at a calcination temperature of 400 °C. They reported that 99.7% of the methyl orange was removed by the ZnO particles. Davari et al. [14] modified ZnO with 6 mol.% Fe₂O₃, achieving a high surface area of 291.35 m²/g and a bandgap energy of 3.38 eV. Using a response surface methodology, they optimized the photocatalytic performance of ZnO/Fe₂O₃ (99.0%) in the presence of a H₂O₂ oxidant. Because of their high photocatalytic activity, the synthesis of ZnO particles has recently been paid much attention.

There are a number of methods for synthesizing ZnO particles, e.g., chemical co-precipitation and the sol–gel, hydrothermal, and organometallic methods [15]. However, such approaches have many disadvantages relating to the use of toxic chemicals during the synthesis of ZnO particles. To minimize the harmful effects of chemicals, green synthesis with an emphasis on plant extracts has been widely used in previous works. Nguyen et al. [16] reported that natural compounds such as polyphenols, quercetin, kaempferol, reducing sugars, gallic acid, and flavonoids extracted from plants or plant tissues can aid in the reduction and stabilization of ZnO. Moreover, green ZnO showed better antimicrobial activity and less toxicity than chemically synthesized ZnO particles [17,18,19]. Indeed, Nguyen et al. [20] demonstrated the multiple functions of green ZnO particles as excellent agrichemical adsorbent photocatalysts against methylene blue dye. Asha et al. [21] showed the biological performance of Piper-longum-mediated ZnO nanoparticles with a hexagonal wurtzite structure in terms of their antibacterial activity against S. aureus and E. coli. In a recent study, Bhattacharjee et al. [22] overviewed the potential of biologically synthesized ZnO nanoparticles for photocatalytic applications. The authors also discussed various optimization methods, such as univariate, multivariate, and linear regression methods, for the removal of pollutants. Among these techniques, chemometrics using mathematical and statistical methods is the most powerful approach to obtain the optimum results. Therefore, the importance of optimization methods should be noted.

The Taguchi orthogonal array (L₉) is considered one of the most effective approaches for multivariate optimization to obtain target responses [23]. This method is primarily used to design an experimental space with appropriate variables for a small number of trials [24]. The Taguchi method also allows one to evaluate the contribution and effect of each operating factor in the experiment [25]. The prediction of results when applying the Taguchi orthogonal array can be conducted through a confirmation test. The coefficient of determination and probability value are calculated to decide whether the model is reliable. Herein, the Taguchi orthogonal array was applied to optimize the photocatalytic degradation efficiency of methylene blue using green ZnO particles. We designed an experimental space of four factors including dye concentration (A), dose (B), time (C), and pH (D), with an array (3⁴) of nine experimental runs. The photocatalytic experiments were independently conducted in batch mode. The green ZnO particles were produced using the flower extract of Chrysanthemum spp. and characterized by several physicochemical techniques. The material was recycled many times and structurally analyzed to assess its stability. To the best of our knowledge, the green synthesis and characterization of green ZnO particles using Chrysanthemum spp. flower extract has not been investigated previously. Furthermore, the application of the Taguchi orthogonal array for the optimization of methylene blue degradation using green ZnO has not been addressed in any publication.

2. Materials and Methods

2.1. Preparation of Chrysanthemum spp. Flower Extract

The fresh flowers of Chrysanthemum spp. plants were collected from the market, washed with distilled H₂O to remove dirt, and then dried in an oven. The dried flowers (2.0 g) were immersed in 100 mL of distilled H₂O and placed into an ultrasound bath (60 W). The ultrasound-assisted extraction of Chrysanthemum spp. flowers was conducted at 60 °C for 60 min. To remove the aqueous extract, the mixture was centrifuged at 6000 rpm for 10 min. The extract solution was stored in a refrigerator at 10 °C.

2.2. Preparation of ZnO Particles Using Chrysanthemum spp. Flower Extract

The synthesis of ZnO particles using the Chrysanthemum spp. flower extract was conducted without the addition of alkaline or other chemicals to obtain completely green targets, following a recent publication [20]. Accordingly, 14.0 g of Zn(NO₃)₂⋅6H₂O (98% purity) was added to 350 mL of Chrysanthemum spp. flower extract. The mixture was completely dissolved and placed in an ultrasound bath (60 W) at 60 °C for 2 h. The liquid was dried at 120 °C to obtain a soft paste. The formation of green ZnO particles was achieved by calcinating the paste sample in a muffle furnace at 600 °C for 60 min. Ultimately, green ZnO particles were collected and stored for subsequent characterization and photocatalytic experiments.

2.3. Characterization of Green ZnO Particles

The surface morphology of the green ZnO particles was investigated by scanning electron microscopy (SEM) using a multi-purpose JSM-6510 scanning electron microscope (JEOL/EO, Tokyo, Japan) equipped with energy dispersive X-ray spectroscopy (EDX) capabilities. The crystal phase structure of the green ZnO particles was examined by X-ray diffraction (XRD) on a D8 Advance Bruker powder diffractometer (Hitachi Inc., Krefeld, Germany) with Cu–Kα radiation (λ = 1.5406 Å) beams as excitation sources. The Fourier-transform infrared spectroscopy (FT-IR) spectra were recorded on a Nicolet 6700 spectrophotometer (Thermo Fischer Scientific Inc., Waltham, MA, USA). An ultraviolet (UV–Vis) spectrophotometer (Shimadzu, Kyoto, Japan) was used to determine the concentration of methylene blue dye at a wavelength of 665 nm.

2.4. Photocatalytic Activities of Green ZnO Particles

Photocatalytic batch experiments involving the green ZnO particles were conducted under solar illumination between 11:00 and 14:30, when ultraviolet irradiation has the strongest intensity. The experimental design was based on the Taguchi orthogonal array to elucidate the interaction between the operating factors. An experimental space comprising 4 factors, including the dye concentration (A), dosage (B), contact time (C), and pH (D), with an array (3⁴) of 9 experimental runs was established, as shown in

Table 1. List of the four operating factors and their levels.

Factor	Unit	Code	Level 1	Level 2	Level 3
Initial dye concentration	mg/L	A	10	30	50
ZnO dosage	g/L	B	0.33	0.67	1.0
Contact time	min	C	30	60	120
pH	-	D	4	7	10

. In detail, the ZnO catalyst at a dosage of 0.33–1.0 g/L was added to 30 mL of methylene blue dye at a concentration of 10–50 mg/L and pH 4–10. The mixture was agitated for 60 min under dark conditions so that an adsorption equilibrium was established. The experimental samples were exposed to sunlight irradiation for 30–120 min. An aliquot of each sample was taken to determine the concentration of methylene blue after the photocatalytic process.

To measure the photocatalytic efficiency of the green ZnO particles, the degradation efficiency RE (%) was defined by Equation (1).

$$RE (\%)=\frac{C_o-C_e}{C_o}\times 100$$

(1)

where C_o, and C_e are the concentrations (mg/L) of methylene blue dye before and after the photocatalytic treatment.

3. Results and Discussion

3.1. Characterization of Green ZnO Particles

In this work, optical analysis was performed to confirm the formation of green ZnO particles synthesized using Chrysanthemum spp. flower extract. Figure 1a shows the UV–visible absorption spectra of Zn²⁺, Chrysanthemum spp. flower extract, and ZnO particles. Accordingly, a maximum absorption peak at 324 nm was observed for the ZnO particles. Figure 1b also indicates a color change from light yellow to brown when Zn²⁺ was added to the Chrysanthemum spp. flower extract under ultrasound irradiation at 60 °C. This phenomenon may be explained by the fact that Chrysanthemum spp. flower extract contains a number of natural compounds such as folic acid, polyphenols, niacin, quercetin, reducing sugars, and flavonoids [26]. These substrates have functional groups that can chelate with Zn²⁺ ions to generate Zn²⁺–natural compound complexes. Under heating and microwave irradiation conditions, the complexes can be converted into ZnO particles. The same maximum absorption value was reported in several recent works that used C. fistula and M. azedarach [27] and Cayratia pedata [28] plant extracts for the bio-mediated synthesis of ZnO particles.

The optical analysis of the green ZnO particles could be better elucidated by UV–visible diffuse reflectance spectroscopy. We observed an absorption edge located between 400 and 420 nm. To determine the bandgap energy of the green ZnO particles, the Tauc equation (Equation (2)) could be used.

$$E_g=\frac{hc}{\lambda }=\frac{1240}{\lambda }$$

(2)

where E_g denotes the bandgap (eV) of the ZnO particles, h denotes Planck’s constant (6.63 × 10⁻³⁴ J·s), λ (nm) denotes the maximum absorption wavelength, and c denotes the velocity of light [29]. The Tauc plot of (αhν)² versus (hν) was established to calculate the bandgap of the ZnO particles. Figure 2 exhibits a bandgap energy of 3.0 eV for fresh ZnO particles synthesized using Chrysanthemum spp. flower extract. This value was well commensurate with the bandgap energy values of both chemically synthesized and bio-mediated ZnO published in the literature [20,30,31]. Therefore, the green ZnO particles could serve as a photocatalyst under solar light illumination, including both ultraviolet and visible light.

Herein, the chemical bonding properties of green ZnO particles synthesized using Chrysanthemum spp. flower extract were determined using FT-IR spectroscopy (

Table 2. Surface chemistry of green ZnO particles synthesized using Chrysanthemum spp. flower extract.

No.	Position	Chemical Bonds	Compounds	Reference
1	3730 cm−1	A medium band of O–H vibrations	Hydroxyl groups of alcohols, polyphenols, polysaccharides, quercetin, genin, and gallic acid in Chrysanthemum spp. flower extract or surface-chemisorbed H2O.	[36]
2	3450 cm−1	N–H vibrations (broad band)	Amide groups, primary amines or proteins from alkaloids, and betalains in Chrysanthemum spp. flower extract.
3	2858–2923 cm−1	C–H stretching (medium band)	Symmetric methylene groups (CH2), aliphatic groups, or aldehydes from natural compounds
4	2385 cm−1	C=C bending (sharp band)	Aromatic compounds, e.g., quinones, coumarins, tannins, or terpenes (such as carotenoids).	[35]
5	2300 cm−1	C–N/C=N (very sharp band)	Tertiary amines or imines.
6	1746 cm−1	C=O stretching (narrow band)	Carbonyl groups of esters, δ-lactone, or ketones.	[34]
7	1632 cm−1	C=C stretching (broad, strong band)	Alkenes.
8	1390 cm−1	C–H bending (strong band)	Aldehydes or reducing sugars.	[33]
9	1124 cm−1	C–O stretching	Secondary or tertiary alcohols.
10	860.1 cm−1	N–H stretching vibrations	Amines.	[33]
11	530–715 cm−1	Zn–O	The formation of zinc oxide nanoparticles using Chrysanthemum spp. flower extract.	[38]

). According to Figure 3, a range of peaks at 530–715 cm⁻¹ was assigned to hexagonal-phase Zn–O bonding, confirming the formation of ZnO particles. The location of these characteristic peaks was in good accordance with previous works [20,27,32]. The existence of primary amines, with N–H stretching vibrations, on the surface of the green ZnO particles could be demonstrated by the typical peak at 860.1 cm⁻¹ [33]. Meanwhile, C–O stretching, diagnosed by a peak at 1124 cm⁻¹, was attributable to secondary or tertiary alcohols. Strong C–H bending footprints, belonging to aldehydes or reducing sugars, were observed at 1390 cm⁻¹ [33]. While alkene compounds could be identified at 1632 cm⁻¹, there was a narrow peak at 1746 cm⁻¹, ascribed to the C=O stretching of carbonyl groups of esters, δ-lactone, or ketones [34]. The strong absorbance intensity of the peak at 2300 cm⁻¹ (C–N/C=N) may have been due to the presence of tertiary amines or imines. At 2385 cm⁻¹, the sharp band of C=C bending may have belonged to aromatic compounds, e.g., quinones, coumarins, tannins, or terpenes (such as carotenoids) [35]. The very typical signals at 2858–2923 cm⁻¹ were due to the absorbance of symmetric methylene groups (CH₂), aliphatic groups, or aldehydes from natural compounds. We observed N–H vibrations (broad band) at 3450 cm⁻¹, which could have signaled amide groups, primary amines or proteins from alkaloids, or betalains in the Chrysanthemum spp. flower extract. Finally, a medium band of O–H vibrations at 3730 cm⁻¹ could be assigned to either hydroxyl groups of alcohols, polyphenols, polysaccharides, quercetin, genin, or gallic acid in the Chrysanthemum spp. flower extract, or surface-chemisorbed H₂O [36]. Several works have reported the same chemical bonds in hexagonal-phase ZnO particles synthesized by bio-based methods using the extracts of various plants, such as Phlomis leaves [32], Cassia fistula and Melia azedarach leaves [27], Canna indica flowers [20], and Myristica fragrans fruit [37]. Therefore, the surface of the hexagonal-phase green ZnO nanoparticles synthesized using Chrysanthemum spp. flower extract contained a variety of functional groups, which were expected to have good photocatalytic activity.

To determine the crystalline structure of the green ZnO particles synthesized using Chrysanthemum spp. flower extract, the X-ray powder diffraction pattern was observed. As shown in Figure 4, the green ZnO particles showed a typical XRD pattern with summits (2θ, degrees) located at 31.92° (100), 34.62° (002), 36.52° (101), 47.72° (102), 56.84° (110), 63.18° (103), and 68.36° (112). Moreover, the pattern profile of the as-synthesized ZnO in this work was matched well with the standard reference for ZnO particles (JCPDS #36–1451), as shown in Figure 4. It was also in line with the pattern profile of the hexagonal wurtzite-structured ZnO reported in previous works [20,27,32]. The pattern of the green ZnO particles showed very-high-intensity peaks, and no strange peaks were observed. This indicated that the green ZnO particles synthesized using Chrysanthemum spp. flower extract had a high degree of purity. To estimate the average size of the green ZnO particles, Scherer’s equation could be used (Equation (3)).

$$D=\frac{K\times \lambda }{\beta \times \cos \left(\theta \right)}$$

(3)

where K is a constant equal to 0.90, β is the full-width at half-maximum height of the diffraction peaks at Bragg’s angle θ, and λ is the wavelength of the X-ray radiation. According to Equation (3), the average size of the green ZnO particles synthesized using Chrysanthemum spp. flower extract was estimated to be 30.24 nm. The particle size of the green ZnO particles in this work was in excellent agreement with results from previous studies using extracts such as Canna indica flowers (29.85 nm) [20], Capparis zeylanica leaves (32 nm) [28], Solanum nigrum leaves (29.79 nm) [39], Costusigneus leaves (31 nm) [40], and Uncaria gambir leaves [41] for the bio-mediated synthesis of ZnO particles.

Figure 5 presents the scanning electron microphotographs of the green ZnO particles synthesized using Chrysanthemum spp. flower extract. The morphology of the green ZnO particles had a level of agglomeration. Due to the calcination during synthesis, crystals grow under the control of many factors, such as van der Waals forces, polarities, and electrostatic attractions [16]. As a result, ZnO particles can form many defects on their surfaces, which results in a large surface area. This phenomenon is explained by the thermal decomposition of natural compounds in Chrysanthemum spp. flower extract, e.g., folic acid, polyphenols, niacin, quercetin, reducing sugars, and flavonoids [26]. The influence of the calcination temperature on the morphology of green ZnO particles was reported by Thi et al. [42]. However, a calcination temperature of 600 °C could be to minimize the self-clustering and agglomeration of green ZnO particles.

The chemical composition of the green ZnO particles synthesized using Chrysanthemum spp. flower extract could be analyzed by energy-dispersive X-ray spectroscopy. According to Figure 6, the ZnO particles contained O and Zn as the main components (>94%) in their structure. Our findings revealed that the ZnO particles had a high degree of purity. A minute amount of other elements such as N (4.23%), Cl (0.29%), P (0.32%), and K (0.96%) was observed in the ZnO nanomaterial. These traces could have come from the Chrysanthemum spp flower extract. The calcination of the sample released most of the volatile components, leaving a number of elements in the structure of the ZnO. Several works have reported similar high-purity ZnO particles synthesized using various plant extracts, such as Hibiscus sabdariffa, Cannabis Jatropha, Tinosporacordifolia, Thymbra spicata, and Cinnamomum camphora [43,44,45,46]. As such, ZnO particles could exhibit sufficient photocatalytic activity against organic dyes. To better elucidate this, the green ZnO particles synthesized using Chrysanthemum spp. flower extract were further investigated.

3.2. Taguchi Design and Model Optimization

In this study, the Taguchi design was used to create an experimental space comprising four factors, including dye concentration (A, 10–50 mg/L); ZnO dosage (B, 0.33–1.0 g/L); contact time (C, 30–120 min); and pH (D, 4–10). As shown in

Table 3. Taguchi L₉ factor and level combinations for assessing the methylene blue degradation efficiency of green ZnO particles.

No.	A (mg/L)	B (g/L)	C (min)	D (pH)	Removal (%)
1	10	0.33	30	4	18.1
2	10	0.67	60	7	68.8
3	10	1.0	120	10	96.9
4	30	0.33	60	10	35.7
5	30	0.67	120	4	48.9
6	30	1.0	30	7	17.6
7	50	0.33	120	7	35
8	50	0.67	30	10	15.8
9	50	1.0	60	4	11.1

, there were nine experiments in total (entry 1–9) for this design, and the average degradation efficiency value (11.1–96.9%) was 38.66%. To better elucidate the effect of the main parameters, the analysis of variance (ANOVA) conducted based on the Taguchi design (3⁴) is presented in

Table 4. ANOVA data for standard average.

Parameter	DF a	SSS b	SEC c	Coefficient	p Value d
A—Initial concentration (mg/L)	1	2476.6	0.193	−1.016	0.006 e
B—Adsorbent dosage (g/L)	1	225.7	0.386	0.613	0.187 f
C—Contact time (min)	1	2692.0	0.084	0.462	0.005 e
D—pH of the solution (−)	1	823.7	1.287	3.906	0.039 e
Total	4	6218.0	–	–	0.009 e

Notes: ^a Degree of freedom, ^b sequential sums of squares, ^c standard error of the coefficient, ^d probability, ^e significant at p < 0.05, ^f not significant at p > 0.05.

.

The p value of the regression was found to be 0.009, corresponding to an F value of 17.38. This value was far lower than 0.05, indicating that the regression model was statistically significant. Moreover, the coefficient of determination R² and adjusted R² were very high at 0.946 and 0.891, respectively. Therefore, the experimental data in this study offered high reliability. The probability values of terms A, B, C, D could be used to clarify whether they were statistically significant. The results revealed that the initial concentration, contact time, and pH were statistically significant at p < 0.05, while the ZnO dose was statistically insignificant. From the ANOVA table, a regression equation was established to show the correlation between the variables and responses, as follows (Equation (4)):

$$\mathrm{Degradation}\mathrm{efficiency}\left(\%\right)=-2.8-1.02\mathrm{A}+18.39\mathrm{B}+0.462\mathrm{C}+3.91\mathrm{D}$$

(4)

To determine the contribution of the parameters, the value of the sequential sums of squares was calculated. The contribution of each parameter was equal to the ratio between the sequential sums of squares of that parameter and the total sequential sums of squares. According to Figure 7, the most important contributor was contact time, and pH was the least important contributor. The contributions were ranked as follows: contact time (rank 1: 43.29%) > initial concentration (rank 2: 39.83%) > pH of the solution (rank 3: 13.25%) > ZnO dosage (rank 4: 3.63%). Importantly, the contact time and initial concentration contributed to 83% of the total main effect. Figure 8a,b show the effect of these parameters on the degradation efficiency; the optimum conditions were obtained at a low concentration, medium dosage, high reaction time, and high pH.

Certain diagnostic plots can be useful for model validation and the assessment of model reliability. Figure 8c provides a visual representation of the data distribution, with red points tending to gather around the blue line. This normal probability plot indicates a high level of compatibility, and there were no abnormal points observed in the dataset. Figure 8d presents the standardized residuals, with all values in the range of −2 to +2. No specific patterns, outliers, or large residuals were observed in the plot. The fitted values were sufficient to ignore the errors caused by the data noise. There was no need to rerun the experiments for adjustment. Therefore, the model was adequate to predict the optimal results, and the assumptions of the regression model were satisfied with high suitability. Model confirmation was conducted to verify the predictions.

3.3. Model Confirmation

Based on

Table 5. Optimal conditions based on the highest signal-to-noise ratio.

No.	Parameters	Optimized Value	Level	Contribution (%) a
1	Initial concentration (mg/L)	10	1	39.83
2	ZnO dosage (g/L)	0.67	2	3.63
3	Contact time (min)	120	3	43.29
4	pH of the solution (-)	10	3	13.25

Note: ^a Predicted value: 99.5%; actual value: 99.0%; relative error: −0.5%.

, the optimal conditions were selected according to the highest signal-to-noise ratio. Accordingly, the optimum value was predicted to be 99.5% when the following conditions were met: initial concentration of 10 mg/L, ZnO dosage of 0.67 g/L, contact time of 120 min, and pH of 10. To verify this prediction, a confirmation test was conducted with the above inputs. We observed that the actual value for the degradation of methylene blue was 99.0%, which was 0.5% lower than the predicated value. This error was minute and acceptable for the model estimation. The Taguchi L₉ (3⁴) orthogonal array design was a reliable approach for the optimization of the photocatalytic degradation of methylene blue dye using green ZnO particles.

3.4. Proposed Mechanism for Dye Degradation

To shed light on the photocatalytic mechanism of the green ZnO particles synthesized using Chrysanthemum spp. flower extract, scavenging experiments were conducted. It was important to measure the degradation efficiency of methylene blue by the green ZnO catalyst in the presence of several scavengers, e.g., ascorbic acid, tert-butanol, and ethylenediaminetetraacetic acid. Because the scavengers reacted with active radicals including hydroxyl (^•OH), superoxide (^•O₂⁻), and hole (h⁺) radicals, the degradation efficiencies were reduced. We observed that the degradation efficiency of the ZnO catalyst in the presence of scavengers was 41.3% for ascorbic acid, 55.8% for ethylenediaminetetraacetic acid, and 88.0% for tert-butanol. Therefore, the contributions of the radicals to the reaction were ranked as follows: ^•O₂⁻ > ^•OH > h⁺. These findings suggested that both ^•O₂⁻ and ^•OH species played a major role in the degradation of methylene blue dye.

Figure 9 describes the proposed mechanisms of the ZnO catalyst. In detail, the ZnO edges initially absorb photons of solar light to generate electrons (e⁻) and holes (h⁺). ^•O₂⁻ radical species can be formed via reactions between electrons adsorbed on the conduction band. ^•OH radical species are generated via a multi-step reaction between holes (h⁺) and H₂O/OH⁻. Methylene blue also adsorbs photons of solar light to generate radicals. The most important reaction between reactive oxygen species (^•O₂⁻, ^•OH) and dye molecules causes the formation of smaller degradation products such as CO₂ and H₂O. The main steps of this plausible photocatalytic reaction are illustrated in Figure 10.

3.5. Recyclability Study

Recyclable catalysts offer many advantages, i.e., reducing the production cost and indirectly minimizing the disposal of pollutants. Therefore, the recyclability of the ZnO catalyst synthesized using Chrysanthemum spp. flower extract for the photodegradation of methylene blue dye in water was investigated. The eluent solvents for washing and desorbing the dye molecules from the catalyst were important; hence, ethanol was selected due to its low cost, effectiveness, and safety [47]. After regeneration, the catalyst was subject to a reaction with methylene blue dye under solar light irradiation. The results shown in Figure 11 indicate that the green ZnO could be reused for five sequential cycles. At an MB concentration of 10 mg/L, the first cycle achieved an efficiency of nearly 97%, while the value of the final cycle remained at 60%, revealing the excellent stability and photocatalytic activity of the green ZnO synthesized using Chrysanthemum spp. flower extract. The significant decrease (~37%) in the degradation efficiency of the green ZnO catalyst might be explained by the gradual deactivation of the catalytic ZnO sites after each reuse cycle. Moreover, the washing process in the regeneration of the ZnO might not have been complete, leaving traces of MB dye bound on the surface of the ZnO. As a result, the photocatalytic activity was reduced after each reuse cycle. The same number of reuse cycles (five) for ZnO was reported in previous works [47,48].

3.6. Stability Study

The stability of the green ZnO synthesized using Chrysanthemum spp. flower extract was assessed using physicochemical analysis. The XRD and FT-IR spectra of the reused and fresh ZnO catalysts after the treatment of methylene blue dye were monitored. As observed from Figure 12, the bandgap energy of the reused ZnO catalyst was 3.0 eV, which was the same as that of the fresh ZnO catalyst. Moreover, Figure 13a shows that some peaks remained for the reused ZnO catalyst. There was also no significant difference in the position of the XRD peaks between the reused and fresh ZnO catalysts. However, the sharp FT-IR peaks at approximately 2300 (C–N/C=N stretching) and 2385 cm⁻¹ (C=C bending) were absent. This could be explained by the fact that the crucial functional groups, i.e., tertiary amines, imines, etc., on the surface of the green ZnO particles were significantly reduced. Therefore, a decreasing trend in the photocatalytic activity of the ZnO particles is indicated in Figure 11. Moreover, minor shifts in other peaks can be observed in Figure 13b, indicating that the washing process of the ZnO might not have been complete, leaving trace amounts of MB dye bound on the surface of the ZnO. However, it could be confirmed that the structure of the ZnO catalyst still had high stability under experimental conditions.

3.7. Comparison Study

To demonstrate the effectiveness of ZnO particles synthesized through various routes, such as bio-mediated and chemical pathways, a comparison study was performed. As shown in

Table 6. Comparison of the photocatalytic degradation of methylene blue dye using ZnO particles synthesized via various pathways.

Method of Synthesis	Light Source	Degradation Efficiency (%)	Catalyst Dose (g/L)	Concentration (mg/L)	Degradation Time (min)	Reference
Bio-mediated (Chrysanthemum spp. flowers)	Sunlight	99.0	0.67	10	120	This work
Bio-mediated (jujube fruit)	Sunlight	85.0	1.0	100	300	[30]
Bio-mediated (Beta vulgaris)	Ultraviolet light, 370 nm	80.0	0.2	5	150	[49]
Bio-mediated (Myristica fragrans)	Ultraviolet light	88.0	0.56	20	140	[37]
Bio-mediated (Pithecellobium dulce peel)	Ultraviolet light, 365 nm	63.0	-	320	120	[50]
Bio-mediated (Justicia spicigera)	Ultraviolet light, Hg lamp 10 W	81.9	1.0	15	90	[52]
Bio-mediated (Suaeda japónica)	Ultraviolet light, 346 nm	54.0	1.0	32	60	[51]
Commercial	Ultraviolet light, Xenon 200 W	66.9	0.5	20	80	[53]
Bio-mediated (Hibiscus sabdariffa)	Ultraviolet light, 10 W	97.0	1.0	15	150	[43]
Bio-mediated (Scutellaria baicalensis)	Ultraviolet light, 365 nm	98.6	0.1	16	210	[54]
Chemical	Ultraviolet light, 10 W	78.5	1.0	15	120	[43]
Chemical	Sunlight	98.0	0.5	100	30	[55]

, the photocatalytic methylene blue degradation of ZnO particles synthesized using Chrysanthemum spp. flower extract produced superior results in terms of degradation efficiency and reaction time compared with other ZnO catalysts. Indeed, the green ZnO particles in this study had a degradation efficiency of 99.0%, which was higher than the values for particles synthesized using plant extracts such as Beta vulgaris [49], Myristica fragrans [37], Pithecellobium dulce [50], and Suaeda japonica [51]. We found that this material had a higher degradation efficiency against methylene blue than chemically synthesized ZnO catalysts reported previously [43]. The degradation reaction in this work was performed under solar light irradiation, with a short reaction time of 120 min. Therefore, ZnO particles synthesized using Chrysanthemum spp. flower extract could be an excellent photocatalyst for the degradation of methylene blue dye in aqueous solution in actual wastewaters.

4. Conclusions

The Taguchi L₉ (3⁴) orthogonal array design was successfully applied to optimize the photocatalytic degradation of methylene blue dye by green ZnO particles synthesized using Chrysanthemum spp. flower extract. The reaction time was determined to be the most important contributor (43.29%) to the methylene blue degradation efficiency. The regression model was statistically significant according to the very low p value of 0.009. The model had a high coefficient of determination, R² = 0.946, which proved the high reliability of the experimental data. The predicted value was 99.5%, which was almost equivalent to the test results at an initial concentration of 10 mg/L, ZnO dosage of 0.67 g/L, contact time of 120 min, and pH of 10. A reaction mechanism was suggested, with the main role being performed by reactive oxygen species (^•O₂⁻, ^•OH). The green ZnO particles could be reused for at least five cycles and exhibited high stability via XRD, FT-IR, and DRS tests. Therefore, green ZnO particles synthesized using Chrysanthemum spp. flower extract could be an excellent photocatalyst for the degradation of organic dyes in real wastewaters.