Sustainable Synthesis of Copper Oxide Nanoparticles: Data-Driven Photocatalysis, Pt-Free Hydrogen Production, and Antibacterial Assessment

Abstract

This study reports the green synthesis of copper oxide nanoparticles (CuO NPs) using Oxystelma esculentum extract as a reducing and stabilizing agent. The state-of-the-art analysis confirmed their spherical morphology, with an average particle size ranging from 20 to 25 nm, while XRD indicated a crystalline structure consistent with the standard JCPDS card. The photocatalytic degradation of norfloxacin (NOR) was optimized using Response Surface Methodology (RSM), which identified the optimal conditions as a reaction time = 47.51 min, CuO-NPs dose = 48.46 mg, NOR dose = 35.90 ppm, and pH = 5.23. Under these optimized conditions, the CuO NPs achieved an initial degradation efficiency of 90%. In addition to photocatalytic degradation, the hydrogen (H₂) evolution performance of the CuO NPs was evaluated, yielding a H₂ production rate of 19.52 mmol g⁻¹ h⁻¹ under visible light. Moreover, the antimicrobial activity of the CuO NPs was assessed, showing significant antibacterial effects with inhibition zones of 8 mm and 9 mm against Klebsiella and Bacillus species. The CuO NPs also exhibited potent anticancer activity with an IC₅₀ value of 15.3 ± 1.40 μM against the HeLa cell line and notable antifungal activity with inhibition rates ranging from 70% to 90% against various fungal species.

1. Introduction

The pervasive presence of fluoroquinolone antibiotics in wastewater from sources like hospitals and industries constitutes a significant environmental and public health threat [1,2]. These persistent pharmaceuticals, along with their metabolites, are frequently detected in rivers, lakes, and even drinking water sources, raising serious concerns about ecological damage and the promotion of antibiotic-resistant bacteria [3,4]. Conventional wastewater treatment methods, including physical separation processes like coagulation and adsorption, are often employed for pollutant removal [5,6]. However, these methods primarily transfer the antibiotics from the aqueous phase to a solid sludge, rather than degrading them. This creates a secondary waste stream that requires further treatment and disposal, and does not eliminate the inherent hazard [7]. Consequently, there is a critical need for advanced, destructive technologies that can mineralize these organic pollutants into less toxic products.

The intense pace of urbanization and industrialization has further worsened this issue, leading to the widespread release of untreated or partially treated wastewater into aquatic ecosystems [8,9,10,11]. These effluents often contains a complex mixture of pollutants, including industrial dyes, toxic heavy metals, and pharmaceutical residues [12]. A significant pathway for antibiotic contamination is through the discharge from medical facilities and the excretion of incompletely metabolized drugs by humans and livestock [4,13,14,15,16]. The persistent presence of these compounds, particularly fluoroquinolones like NOR and ciprofloxacin, poses severe risks, such as disrupting essential microbial communities involved in nutrient cycling (e.g., the nitrogen cycle) [17,18], promoting the development of antibiotic-resistant bacteria [19] and facilitating their bioaccumulation within the food web [20,21,22,23,24].

Conventional wastewater treatment methods often fall short of mitigating this issue [17]. Physical and chemical processes, such as coagulation, filtration, and adsorption, are primarily separation techniques that transfer antibiotics from the water phase to a solid sludge, rather than degrading them [25]. This not only fails to eliminate the inherent hazard but also generates a concentrated secondary waste stream that requires further treatment and disposal [25,26]. While biological treatment can degrade some organic pollutants, its efficacy is limited by slow reaction rates, specific operational requirements, and the inherent resistance of many complex antibiotic molecules to microbial breakdown under standard aerobic conditions [27,28,29].

Among all the techniques, photocatalytic degradation is considered an adaptable technique to mitigate the organic contaminants. The reaction in this approach proceeds with the absorption of light by the photocatalyst surface, resulting in the chemical modification of the pollutants to their less hazardous reactive intermediates [30]. In contrast to homogeneous and heterogeneous catalysis, heterogeneous catalysis prevails due to its potential to degrade the antibiotic/dye to CO₂ and H₂O completely [31,32,33]. Moreover, heterogeneous catalysis holds several advantages, such as recyclability, chemical stability, non-toxicity, and pleasant separation [34]. Over the years, several classes of metal oxides, such as MgO, WO₃, SnO₂, ZnO, CeO₂, TiO₂, V₂O₅, and Bi₂O₃, have been acknowledged for their beneficial potential for dye and antibiotic degradation [35,36,37,38,39]. These metal oxides are considered ideal photocatalysts because of their auspicious electronic structure, charge transfer features, light absorption characteristics, and potential to provide charge carriers in the photocatalytic system [40]. Additionally, they hold a convenient stoichiometric composition and are bio-coherent.

Various physical and chemical techniques have been utilized for the preparation of CuO nanoparticles, including sol–gel, co-precipitation, hydrothermal, thermal decomposition, and electrochemical routes. While these methods often yield uniform and crystalline products, they typically require high energy input, costly precursors, or toxic reducing/stabilizing agents. In contrast, biologically assisted syntheses—using plant extracts, microorganisms, or biopolymers—have emerged as greener alternatives that offer simplicity, low toxicity, and scalability. A concise comparison of representative CuO synthesis methods is presented in Table S6, highlighting the eco-friendly and sustainable aspects of the present work.

In the context of this work, the term ‘green synthesis’ specifically refers to the utilization of plant-derived phytochemicals as harmless reducing and stabilizing agents, replacing conventional toxic chemicals, and the use of water as a solvent, which collectively minimizes the environmental footprint of the nanoparticle fabrication process.

This study presents a comprehensive investigation into the multi-functional applications of copper oxide nanoparticles synthesized via a novel green route using Oxystelma esculentum leaf extract. The primary aim of this work is to synthesize these CuO nanostructures and systematically evaluate their potential in three key areas: (1) the visible-light-driven photocatalytic degradation of the antibiotic NOR, (2) photocatalytic H₂ production via water splitting, (3) broad-spectrum biological activity against bacterial and fungal pathogens, as well as anticancer activity.

The novelty of this work lies in its integrated and data-driven approach. It is the first report, to our knowledge, to utilize Oxystelma esculentum for the sustainable synthesis of CuO nanostructures. Furthermore, the work is distinguished by its methodological rigor in optimizing the photocatalytic process using Response Surface Methodology (RSM), providing a robust statistical model for industrial scalability. Finally, the concurrent demonstration of significant efficacy in environmental remediation (antibiotic degradation, H₂ production) and biomedical applications (antimicrobial, anticancer) from a single, green-synthesized nanomaterial underscores its versatility and practical potential.

2. Characterization of the Nanostructures

2.1. Zeta Potential

Zeta potential was measured to determine the surface charge and assess the electrostatic stability of the synthesized nanostructures in an aqueous dispersion. A Nano ZS zetasizer system (Malvern Instruments, Malvern, UK) was utilized for the experiment. For the measurement, a 0.25 mg/mL dispersion of the nanostructures was prepared in deionized water and analyzed. The spectrum shown in Figure 1A peaks at −23.5 mV with a zeta deviation of 7.76 mV. The observed peak breadth suggests a distribution of surface charges, indicating heterogeneity within the nanoparticle population. This could arise from slight variations in particle size or the non-uniform distribution of capping agents from the plant extract. A zeta potential magnitude of −23.5 mV typically confers moderate colloidal stability, preventing immediate aggregation but suggesting the dispersion may not be stable over very long periods. The conductivity of the dispersion medium was 0.162 mS/cm. This value provides context for the solution’s ionic strength, which can compress the electric double layer around the nanoparticles and influence the measured zeta potential [41].

2.2. Fourier Transform Infrared Spectroscopy (FTIR)

FTIR analysis of the as-synthesized nanostructures was conducted to determine the presence of essential functional groups that primarily serve as the nanostructures’ reducing and capping agents [42,43,44]. The results, as provided in Figure 1B, cover a broad range from 400 to 3500 cm⁻¹. The peak at 3304.84 cm⁻¹ is associated with O–H stretching vibrations, indicative of hydroxyl groups from adsorbed water or residual phytochemicals. The peaks at 1326.87 and 1021.48 cm⁻¹ correspond to C–O stretching vibrations, typically observed in alcohols, phenols, and esters. The peak at 829.00 cm⁻¹ corresponds to C=C stretching, characteristic of alkenes. These organic functional groups (O-H, C-O, C=C) likely originate from the organic compounds in the Oxystelma esculentum extract, which served as reducing and capping agents during the synthesis.

Despite calcination at 400 °C, these discernible bands suggest that a thin layer of organic capping agents remains on the nanoparticle surface. This is a common characteristic of green-synthesized nanoparticles and can contribute to their colloidal stability and influence their surface reactivity. Peaks in the 400–500 cm⁻¹ region are characteristic of Cu–O stretching vibrations, which confirm the successful formation of CuO nanoparticles and the presence of a well-formed crystalline structure [45,46].

2.3. X-Ray Diffraction (XRD)

The crystallinity of the synthesized nanostructures was evaluated using powder XRD analysis, and the XRD spectrum of CuO nanostructures is presented in Figure 1C. The major peaks observed at 2θ values of 32.496°, 35.495°, 36.487°, 38.73°, 38.957°, 46.248°, 48.725°, 50.671°, 53.451°, 58.335°, 61.533°, 65.785°, 66.248°, 67.943°, 68.089°, 72.427°, 74.287°, and 75.225° correspond to the hkl values of −110, 002, −111, 111, 200, −112, −202, 112, 020, 202, −113, 022, −311, 113, −220, 311, 004, and −222, respectively. These planes are well-aligned with the monoclinic phase of CuO, as referenced in JCPDS card No. 00-045-0937, confirming that CuO is the dominant crystalline phase in the synthesized material. Additionally, a low-intensity peak at 2θ ≈ 43.561° was identified, corresponding to the cubic phase of Cu₂O (JCPDS card No. 01-077-0199). This minor Cu₂O phase is likely formed due to partially reducing Cu²⁺ ions during the green synthesis process, which utilizes plant extract containing complex reducing agents.

The XRD spectrum demonstrates distinct and sharp peaks, indicating the high crystallinity of the CuO nanoparticles [47]. The dominant monoclinic crystal system and the CuO structure’s centrosymmetric nature highlight the well-defined arrangement. Centrosymmetric crystals, with their inversion symmetry, are known to influence optical and electronic properties, making them suitable for various functional applications. The presence of the minor Cu₂O phase does not detract from the primary functional properties of CuO, as evidenced by the photocatalytic, antibacterial, and H₂ production experiments presented in this study. Moreover, the average crystallite size of the CuO nanoparticles was estimated using the Scherrer equation, D = Kλ/(β cosθ), where D is the crystallite size, K is the Scherrer constant (0.9), λ is the X-ray wavelength (0.15406 nm for Cu Kα), β is the full width at half maximum (FWHM) in radians, and θ is the Bragg angle. The most intense peak at 2θ = 35.495° (002 plane) was used for the calculation. The calculated average crystallite size was approximately 19.25 nm, which is consistent with the nanoscale dimensions of the synthesized material.

2.4. Scanning Electron Microscopy (SEM)

Scanning electron microscopy unveiled the structural and morphological characteristics of CuO nanostructures engineered from the Oxystelma esculentum. The SEM images revealed that the synthesized CuO nanoparticles predominantly exhibit a spherical morphology. The images show a degree of aggregation, which is typical for dry nanopowders due to van der Waals forces and the sample preparation process for SEM. The nanoparticles were well-dispersed with minimal aggregation, which can be attributed to the stabilizing effect of the phytochemicals present in the plant extract. The primary particle size, as indicated by XRD, AFM, and TEM analyses, was in the nanoscale range. This observation highlights the successful synthesis of CuO nanostructures and underscores their potential for various environmental applications. SEM images of the synthesized CuO nanostructures are presented in Figure 2A–F. On the other hand, the elemental composition and the corresponding elemental data obtained from the extended EDX microscope are presented in Figure S1 and Tables S1 and S2. From the spectrum, the peaks of Cu and O are apparent, and the same percentage composition of the nanostructures is a clear indication of the prosperous synthesis of CuO nanostructures.

2.5. Atomic Force Microscopy (AFM)

Atomic force microscopy provides a complementary perspective for characterizing metallic nanostructures. Through its high-resolution imaging potential, AFM explores the nanoscale topography of nanostructures, fostering valuable insights into their structural and surface characteristics. The fine details of particle size, shape, and surface roughness can be discerned with remarkable precision through AFM analysis. Figure 3A–D below presents the AFM images of the prepared nanostructures. The AFM images of the CuO NPs revealed a distinct surface morphology characterized by smoothness with occasional ridges. The particles exhibit a predominantly spherical shape, consistent with the expected morphology of the synthesized nanoparticles. The distribution of the nanoparticles across the surface was uniform, with some degree of isolation. The AFM analysis also provided detailed height profiles, from which the average particle size was determined to be approximately 9.82 nm. This size, measured from individual particles, is smaller than the agglomerates observed in SEM. The observed size distribution ranged from 1.4 nm to 12.6 nm, indicating a population of primary nanoparticles. The observed morphology, particularly surface roughness and particle size, is expected to significantly affect the performance of the prepared nanoparticles in photocatalysis and antimicrobial applications. The rougher surface with higher RMS (root-mean-square) values suggests a greater number of active sites, which could enhance NOR degradation under visible light. Furthermore, the relatively small, uniformly distributed particles are likely to contribute to efficient H₂ evolution and antimicrobial activity.

2.6. Transmission Electron Microscopy (TEM) and SAED Analysis: Resolving Primary and Secondary Particle Size

TEM and SAED analysis were employed to provide a definitive understanding of the morphology and crystallinity of the synthesized CuO nanoparticles. This analysis is crucial for reconciling the particle sizes obtained from XRD, AFM, and SEM, as each technique probes a different structural level of the nanomaterial. The low-magnification TEM images (Figure 4A,B) reveal that the CuO nanoparticles form larger aggregates ranging from 20 to 100 nm. This observation is consistent with the aggregated structures observed in SEM and explains the larger size range reported there. The formation of these secondary aggregates is typical for dry nanopowders due to van der Waals forces. Figure 4C provides a more detailed view of the surface texture, suggesting a granular structure composed of smaller primary particles. The HRTEM images (Figure 4D,E) provide critical insight into the primary particle size, showing well-defined lattice fringes that confirm the material’s high crystallinity. The measured lattice spacing corresponds to the crystallographic planes of monoclinic CuO. The size of these individual crystalline domains visible in HRTEM is approximately 5–10 nm. The SAED pattern (Figure 4F) shows distinct diffraction rings, confirming the polycrystalline nature of the aggregates and indexing to the (002), (111), and (202) planes of CuO [47].

2.7. X-Ray Photoelectron Spectroscopy (XPS)

XPS of the prepared nanostructures was performed to evaluate the valence electrons, elemental composition, chemical states, and electronic configurations of the elements present. The XPS micrographs (survey spectrum, Cu 2p, and O 1s) are provided in Figure 5A–C below. The survey spectrum comprises three major peaks: Cu 2p (933.17 eV), O 1s (529.47 eV), and C 1s. The appearance of the C1s peak is typically due to adventitious carbon or organic residues from the synthesis process. The spectrum of Cu displays the prevalence of two major peaks present at 933.17 eV and 953.55 eV, which correspond to the presence of Cu in the form of Cu 2p_3/2 and Cu 2p_1/2 having spin–orbit coupling with the conventional energy of 20 eV [48]. Continuing this, two more peaks also appeared in the same spectrum at 943.5 eV and 962.5 eV, which had relatively higher binding energies than the main spin–orbit components, which correlate with the satellite peak of Cu²⁺ and hint at the partially filled d-block (3d9) of Cu. On the other hand, the O 1s peak at around 529.47 eV is characteristic of oxygen bonded to copper in the CuO structure. Additionally, a second peak in the spectrum of O1s present at 533 eV illustrates the occupancy of absorbed H₂O on the surface of nanostructures [49]. Hence, these results encourage the presence of Cu and O in the lattice of CuO.

2.8. Photocatalytic Activity Against NOR

2.8.1. CCD and Statistical Analysis

The RSM-CCD-designed standard table for photocatalytic NOR degradation is presented in

Table 1. RSM-CCD designed a standard DOE table for the photocatalytic NOR degradation.

Runs	Variables								D%
	Actual Variables				Coded Variables
	Reaction Time (min)	CuO-NPs Dose (mg)	NOR Dose (ppm)	pH (Units)	A (min)	B (mg)	C (ppm)	D (units)
1	60	20	45	10	+1	−1	+1	+1	74.23
2	60	20	45	5	+1	−1	+1	−1	79.99
3	30	20	20	10	−1	−1	−1	+1	91.63
4	45	35	32.5	7.5	0	0	0	0	87.6
5	30	50	20	5	−1	+1	−1	−1	86.19
6	45	35	32.5	7.5	0	0	0	0	91.07
7	45	35	32.5	7.5	0	0	0	0	90.34
8	60	50	45	10	+1	+1	+1	+1	76.87
9	60	20	20	10	+1	−1	−1	+1	86.08
10	60	50	20	10	+1	+1	−1	+1	87.18
11	45	65	32.5	7.5	0	+$\alpha$	0	0	82.2
12	60	50	45	5	+1	+1	+1	−1	94.26
13	45	35	32.5	2.5	0	0	0	-$\alpha$	93.86
14	45	5	32.5	7.5	0	-$\alpha$	0	0	69.98
15	45	35	57.5	7.5	0	0	+$\alpha$	0	71.1
16	30	20	45	10	−1	−1	1	1	73.51
17	45	35	7.5	7.5	0	0	-$\alpha$	0	79.59
18	60	50	20	5	+1	+1	−1	−1	85.75
19	45	35	32.5	7.5	0	0	0	0	92.67
20	45	35	32.5	12.5	0	0	0	+$\alpha$	88.73
21	30	50	20	10	−1	+1	−1	+1	82.38
22	45	35	32.5	7.5	0	0	0	0	90.05
23	45	35	32.5	7.5	0	0	0	0	92.49
24	30	20	20	5	−1	−1	−1	−1	78.99
25	75	35	32.5	7.5	+$\alpha$	0	0	0	84.15
26	30	50	45	5	−1	+1	+1	−1	91.12
27	30	50	45	10	−1	+1	+1	+1	77.56
28	60	20	20	5	+1	−1	−1	−1	79.32
29	30	20	45	5	−1	−1	+1	−1	84.12
30	15	35	32.5	7.5	−$\alpha$	0	0	0	84.62

. The order of the experiments, as suggested by the standard table, was randomized to minimize experimental errors. The acquired experimental data for photocatalytic efficacy ranged from 69.98% to 94.26%, indicating a maximum-to-minimum ratio of 1.34. Acquiring a ratio value below the transformation threshold of 10 suggested that no transformation was needed before starting model formation/testing on the acquired dataset. The model analysis resulted in the formation of the following coded quadratic equation, accurately representing the photocatalytic NOR degradation, as follows:

$$\widehat{\mathrm{P}}\%=+90.70-0.115\mathrm{A}+2.41\mathrm{B}-1.79\mathrm{C}-1.69\mathrm{D}+0.9650\mathrm{A}\mathrm{B}-0.0062\mathrm{A}\mathrm{C}+0.0238\mathrm{A}\mathrm{D}+1.41\mathrm{B}\mathrm{C}-2.27\mathrm{B}\mathrm{D}-4.02\mathrm{C}\mathrm{D}-1.36{\mathrm{A}}^2-3.44{\mathrm{B}}^2-3.62{\mathrm{C}}^2+0.3638{\mathrm{D}}^2$$

(1)

The coded equation can be used to perform predictive analysis and obtain the P% values for individual parameters. Positive coefficients (including B, AB, AD, BC, D²) positively reinforce the P% values. In contrast, the negative coefficients (including A, C, D, AC, BD, CD, A², B², and C²) harm the P% values. Moreover, the larger the numerical coefficient value (such as B, C, D, BC, BD, CD, A², B², and C²), the stronger its correlation with the P% results [50]. The acquired quadratic model indicated that the variables were highly correlated, suggesting the need for a second-order polynomial model, as the conventional kinetic approach cannot account for these correlations. The statistical parameters associated with the quadratic model are also presented in

Table 2. Statistical parameters are associated with the quadratic model.

Model	p-Value	R2	Predicted R2	Adjusted-R2	Lack-of-Fit p-Value
Quadratic	$<$0.0001	0.9541	0.7897	0.9113	0.3948

The validity and statistical significance of the elected model are highlighted by the acquisition of the p-value $<$ 0.0001. Conventionally, the acquisition of a p-value $<$ 0.05 is needed to consider the model statistically significant at the confidence interval value of 95%. The lower p-values for the quadratic model suggested that it was applicable even at a 99% confidence interval. The statistically insignificant p-values of 0.2028, 0.1729, and 0.9742 observed for the linear, 2FI, and cubic models revealed that these models were unsuitable for explaining the acquired results. The parametric analysis of the understudy models is provided in the Supplementary Materials.

2.8.2. Model Testing and Statistical Analysis

The model testing associated with the acquired results presented in Table 2 of the manuscript is presented in

Table 3. Model testing for photocatalytic NOR degradation via CuO-NPs.

Model Fit Summary
Source	Standard deviation	Sequential p-value	Lack of Fit p-value	R2	Adjusted R2	Predicted R2	PRESS	Remarks
Linear	6.65	0.2028	0.0032	0.2049	0.0776	−0.14788	1596.38	Not suggested
2-factor interaction (2FI)	6.15	0.1729	0.0041	0.4838	0.2120	0.13292	1205.86	Not suggested
Quadratic	2.06	<0.0001	0.3948	0.9541	0.9112	0.78973	292.42	Suggested
Cubic	2.69	0.9742	0.0684	0.9634	0.84857	−2.4812	4841.35	Aliased
Sequential Model Sum of Squares [Type-1]
Source	Sum of Squares		Degrees of freedom	Mean Square	F-value	p-value		Remarks
Mean vs. Total	2.130 × 105		1	2.130 × 105				--
Linear vs. mean	284.92		4	71.23	1.61	0.2028		Not suggested
2FI vs. Linear	387.85		6	64.64	1.71	0.1729		Not suggested
Quadratic vs. 2FI	654.14		4	163.53	38.44	<0.0001		Suggested
Cubic vs. Quadratic	12.98		8	1.62	0.2234	0.9742		Aliased
Residual	50.83		7	7.26				--
Total	2.144 × 105		30	7145.15				--
Lack of fit tests
Source	Sum of Squares		Degrees of freedom	Mean Square	F-value	p-value		Remarks
Linear	1088.41		20	54.42	15.65	0.0032		--
2FI	700.56		14	50.04	14.39	0.0041		--
Quadratic	46.42		10	4.64	1.34	0.3948		Suggested
Cubic	33.45		2	16.72	4.81	0.0684		Aliased
Pure Error	17.38		5	3.48				--

. The models encompassing linear, 2FI, quadratic, and cubic models were implemented on the dataset, and the validity of each model was assessed on an individual and combinational basis.

2.8.3. ANOVA Analysis

The validity of the developed quadratic model for explaining the acquired results is investigated by performing the ANOVA analysis presented in

Table 4. ANOVA analysis.

Source	Sum of Squares	Degree of Freedom	Mean Square	F-Value	p-Value
Model	1326.90	14	94.78	22.28	<0.0001
A	0.3174	1	0.3174	0.0746	0.7885
B	139.59	1	139.59	32.82	<0.0001
C	76.47	1	76.47	17.98	0.0007
D	68.55	1	68.55	16.11	0.0011
AB	14.90	1	14.90	3.50	0.0809
AC	0.0006	1	0.0006	0.0001	0.9905
AD	0.0090	1	0.0090	0.0021	0.9639
BC	31.58	1	31.58	7.43	0.0157
BD	82.63	1	82.63	19.42	0.0005
CD	258.73	1	258.73	60.82	<0.0001
A2	51.01	1	51.01	11.99	0.0035
B2	324.11	1	324.11	76.19	<0.0001
C2	360.18	1	360.18	84.67	<0.0001
D2	3.63	1	3.63	0.8532	0.3703
Residual	63.81	15	4.25
Lack of Fit	46.42	10	4.64	1.34	0.3948
Pure Error	17.38	5	3.48
Cor Total	1390.71	29

. The quadratic model exhibited a p-value of $<$0.0001, revealing that there is a possibility of only a 0.01% chance that this p-value is achieved owing to noise. It is a well-known fact that the higher the numerical value of the F-value, the higher the significance of that parameter in developing the model. The parameter significance concerning the F-value was found to be C² $>$ B² $>$ CD $>$ B $>$ BD $>$ C $>$ D $>$ A² $>$ BC $>$ AB $>$ D² $>$ A $>$ AD $>$ AC. As observed, the NOR amount and CuO-NPs were the two most significant factors determining the photocatalytic efficacy of the model reaction. Apart from acquiring a significant p-value for the quadratic model, the significance of the individual model terms was also investigated. The terms B, C, D, BC, BD, CD, A², B², and C² were significant, with a p-value $<$ 0.05. The acquisition of considerable interaction terms of BC, BD, and CD highlighted that the interaction terms need to be included for proper modeling of the photocatalytic reaction. Therefore, RSM-based contour mapping is required to be studied to investigate the interaction terms of the parameters [51].

The diagnostic tests, including standard probability plots, predicted vs. experimental plots, and residual vs. predicted plots, were also investigated for the developed quadratic model, as presented in Figure 6A. The acquisition of concentrated data points along the diagonal of the predicted P% versus actual P% plot (Figure 6A) reveals an extremely high correlation of 95.41% for the understudy experiment. Furthermore, the lower experimental error percentage of 04.59% further affirmed that the reactions were performed with acceptable scientific accuracy and that the quadratic model was valid for the acquired experimental results [52]. Figure 6B illustrates the normality of the acquired data. The closeness of the data points to the diagonal affirms that any data transformation was unnecessary in the understudied case [53]. Figure 6C also represents the residual vs. predicted P% plot. The acquisition of data points within the threshold suggests that the predicted values were accurate and precise enough to reproduce the experimental results of photocatalytic NOR degradation.

2.8.4. Contour and RSM Plotting

The contour and RSM mapping associated with the interaction terms are presented in Figure 7. It should be mentioned here that among the interaction terms (AB, AC, AD, BC, BD, and CD), only the statistically significant interaction terms, including BC, BD, and CD, were utilized as the contributions of the insignificant interaction terms (possessing a p-value $>$ 0.05) are negligible for the working model. The contour and RSM mapping of the interaction terms present the same information. The acquisition of contours rather than straight lines in the contour maps of BC (Figure 7A), BD (Figure 7C), and CD (Figure 7E) confirms the existence of a strong correlation between P% and the interaction terms. The RSM plot of BC (CuO-NPs dose $\times$ NOR dose; Figure 7B) indicated that the optimum concentrations of both factors were required for achieving the maximum value of degradation (indicated by the colored region of the map). At lower concentrations of CuO-NPs, the NOR degradation was found to be relatively lower, followed by the attainment of maximum degradation in the range of 38–42 mg. Further increments in the catalyst do not impact the degradation. The NOR degradation was reduced by increasing the catalyst dose from 42 mg. This negative impact can be attributed to enhanced recombinant electron–hole pair reactions in the medium, driven by extremely high concentrations of CuO-NPs [54]. The lower NOR dose was the more effective concentration for achieving maximum degradation. Lower degradation was observed at high pollutant doses, as the CuO-NP dose was insufficient to degrade effectively under these conditions. The RSM plot of BD (CuO-NPs $\times$ pH; Figure 7D) indicated that the high concentration of the CuO-NPs was essential for the degradation under the fixed NOR concentration. Moreover, the lower pH values were more effective in achieving maximum degradation. The RSM plot of CD (NOR dose $\times$ pH; Figure 7E) affirms that under acidic conditions, the maximum P% degradation was achieved for all the NOR concentrations studied. These results showed that acidic conditions were highly conducive to NOR degradation.

2.8.5. Optimization

Ramp plots of optimized reaction parameters under the defined set of conditions are presented in Figure 8. At first, the software was given the criterion of maximizing the P% values in the case of NOR with the objective significance of +++++, as presented in Figure 8A. The optimized reaction parameters with the maximum predicted P% value of 94.34% were found to be reaction time = 47.51 min, CuO-NPs dose = 48.46 mg, NOR dose = 35.90 ppm, and pH = 5.23 units. The acquired experimental value for this reaction was found to be 95.21%, indicating that the point prediction test performed for the developed quadratic model was highly effective in optimizing the P% values for NOR degradation. Moreover, there are a few desirable industrial criteria, including complete 100% degradation, higher NOR degradation, and lower catalyst usage with the significance level of +++, which were also tested, as presented in Figure 8B. The following optimized reaction parameters, including reaction time = 43.24 min, CuO-NPs dose = 32.29 mg, NOR dose = 43.53 ppm, and pH = 5 units, generate the predictive value of 90.74%, which was not much lower than the above reaction. This information is beneficial for deciding the reaction conditions at the industrial scale while keeping the economic aspect of the reaction in mind.

2.8.6. Recyclability Test of the Synthesized Photocatalyst

The recyclability of the synthesized photocatalyst was evaluated by conducting multiple cycles of NOR photodegradation. The photocatalyst’s performance was monitored over five consecutive cycles under identical experimental conditions to assess its stability and efficiency [55]. In the first cycle, the photocatalyst demonstrated an impressive degradation efficiency of 90%, effectively reducing the NOR concentration in the solution. Upon completing the first cycle, the catalyst was recovered by centrifugation, washed with ethanol and water, and reused for subsequent cycles. During the second cycle, the photocatalyst maintained a high degradation efficiency of 88.5%, indicating only a slight decrease in activity. This trend continued with the third cycle, where the degradation efficiency was recorded at 85%. The marginal reduction in efficiency across cycles suggests that the photocatalyst retained most of its photocatalytic activity. In the fourth cycle, the photocatalyst still exhibited a substantial degradation efficiency of 82.5%. After five cycles of use, the photocatalyst achieved a photodegradation efficiency of 78%, which, while lower than the initial cycle, still represents a significant activity level. These results demonstrate the photocatalyst’s strong recyclability and durability, retaining 78% of its original degradation efficiency after five cycles. The slight decrease in efficiency across cycles could be attributed to minor structural changes in the catalyst. However, the overall performance indicates that the catalyst is robust and can be effectively reused multiple times without significant loss of activity. The graphical illustration of the recyclability test is provided in Figure S2.

2.8.7. Interpretation of the Photocatalytic NOR Removal Results

The interpretation of the photocatalytic results in the NOR degradation can be done as follows:

Elaboration of the Results of RSM

The present study involves the use of RSM to optimize the NOR degradation by CuO nanoparticles. RSM was employed to optimize the photocatalytic NOR degradation by evaluating the effects of various parameters (e.g., pH, catalyst dosage, initial NOR concentration, and reaction time). This statistical tool helps in understanding the interactions between these variables and in determining the optimal conditions for maximum degradation efficiency. The RSM analysis typically involves an ANOVA (Analysis of Variance) test, which identifies the factors that significantly affect the degradation efficiency. For instance, pH and catalyst dosage might show a strong influence on the degradation rate, while other factors like light intensity or initial concentration could have less impact. The response surface and contour plots generated by RSM visually represent the interaction between two variables while keeping others constant. These plots help identify the optimal conditions for NOR degradation. This methodology provides a set of optimal conditions predicted to yield the highest degradation efficiency. The predicted results were then compared with the actual experimental results to validate the model. Moreover, it often reveals whether the combined effects of certain factors are synergistic (enhancing degradation) or antagonistic (reducing degradation). For example, a synergistic interaction between pH and catalyst dosage might suggest that adjusting both factors together enhances the generation of reactive oxygen species (ROS). The optimized conditions derived from RSM provide a foundation for scaling up the photocatalytic process. The optimized values obtained from this approach can be applied in real-world scenarios, such as wastewater treatment plants. The obtained values of reaction parameters for photocatalytic NOR degradation by CuO nanoparticles are: reaction time = 43.24 min, CuO-NPs dose = 32.29 mg, NOR dose = 43.53 ppm, and pH = 5. Although the optimum pH (≈ 5.2) is below the point of zero charge of CuO (pHₚzc ≈ 9), and both the CuO surface and NOR molecules are positively charged, efficient degradation was still achieved. This indicates that non-electrostatic interactions, such as hydrogen bonding, π–π stacking, and surface complexation, together with ROS-driven oxidation (^•OH, ^•O₂⁻, and ¹O₂) play dominant roles in the adsorption and degradation process rather than simple electrostatic attraction.

Kinetics Studies

The kinetics studies of the photocatalytic NOR degradation were examined by the implementation of the Langmuir-Hinshelwood (L-H) model [56,57]. This model is convenient as it typically computes the monolayer sorption of degradation products and reactants involved in the reaction on the photocatalyst surface. In photocatalytic degradation, the reaction rate is determined by a “rate-controlling step.” The mathematical illustration of this model can be done as follows:

$$\mathrm{r}=\left(-{\displaystyle \frac{\mathrm{d}\mathrm{C}}{\mathrm{d}\mathrm{t}}}\right)=(\mathrm{k}\mathrm{K}\mathrm{C})/(1+\mathrm{K}\mathrm{C})$$

(2)

where the rate of reaction (r) originally relies on the NOR concentration (C), the reactant’s adsorption equilibrium constant (K), and the specific reaction rate constant (k). The graphical illustration of the L-H model is provided in Figure S3, and the calculated rate constant for the prepared photocatalyst was found to be 0.0206 min⁻¹. This enhanced value of the rate constant is directed at the superior photodegradation capability of the synthesized CuO nanoparticles and can be correlated to the availability of more active sites and improved surface charge separation by the prepared nanocomposites. This intriguing phenomenon also corresponds to increased ROS production, which are known to expedite NOR photodegradation.

Mechanism of Photocatalytic Degradation and Effect of Radical Scavengers

It is well known that photocatalytic degradation initiates with the initial exposure of the photocatalytic system to light [58,59,60]. Upon exposure, they absorb photons with greater energy than or equal to their band gap. This energy excites electrons (e⁻) from the valence band to the conduction band, leaving behind holes (h⁺) in the valence band. The separation of electron-hole pairs is the key to initiating the photocatalytic reactions. The electrons in the conduction band (CB) and the holes in the valence band (VB) are highly reactive species that drive redox reactions. The excited electrons can transfer to molecular oxygen (O₂) adsorbed on the surface of the CuO NPs, reducing it to superoxide anions (O₂^•−). These superoxide radicals are highly reactive and can attack the NOR molecules, leading to their degradation. Meanwhile, the holes in the valence band can oxidize water molecules (H₂O) or hydroxide ions (OH⁻) present in the solution to generate hydroxyl radicals (^•OH), which are among the most powerful oxidizing agents in photocatalysis. The hydroxyl radicals (^•OH) and superoxide anions (O₂^•−) generated on the CuO NPs surface attack the NOR molecules, breaking down their complex structure into smaller, less harmful products. The photocatalytic process involves cleavage of the fluorine and piperazine rings, leading to NOR degradation into intermediate products and ultimately mineralization into CO₂, H₂O, and other inorganic ions.

Thus, to extend the examination of the characteristic leading role of these ROS in photocatalytic degradation, different scavengers were employed to determine the hindrance caused by these radical quenchers. The experiment used four different types of scavengers: benzoquinone, methanol, sodium azide, and ethylenediaminetetraacetic acid. The results of the impact of different scavengers on photocatalytic degradation are provided in Figure S4. Out of all these, most of the hindrance was observed in the addition of sodium azide, as a mere 23% of the degradation occurred. In addition, the trend of inhibition continued with methanol, BQ, and EDTA, with allowed degradation of 25.5, 33, and 65%, respectively. These results indicate that out of all ROS, holes and hydroxyl radicals play a significant role, leaving behind electrons and superoxide radicals.

The optimal degradation efficiency at pH ~5.23 is a critical finding. While this pH suggests electrostatic repulsion between the positively charged CuO surface and the protonated NOR molecule, the observed high efficiency confirms a strong, favorable interaction. This apparent paradox indicates that electrostatic forces do not primarily govern the adsorption but is dominated by stronger, specific interactions that effectively overcome the Coulombic repulsion. These interactions include:

• π-π stacking/cation-π interaction: Interaction between the electron-rich NOR aromatic rings and the positively charged catalyst surface or residual carbon from the plant extract.
• Hydrogen bonding: between the carbonyl (-C=O), ketone, and carboxylic acid groups of NOR and the surface hydroxyl groups (-OH) of CuO.
• Surface complexation: coordination of the ketone and carboxylate oxygen atoms of NOR with the Lewis acid sites (copper cations) on the catalyst surface.

This multi-faceted binding mechanism ensures the close proximity of NOR to the catalyst surface, which is essential for its efficient degradation by photogenerated ROS. The radical scavenger tests confirm that the degradation itself is primarily mediated by holes (h⁺) and hydroxyl radicals (^•OH), which oxidize the pre-adsorbed NOR molecules.

2.8.8. Importance and Relationship of the Synthesized Nanoparticles on the Synergistic NOR Removal

The elaboration on the significance and relationship of the ascertained morphology of the prepared CuO nanostructures to their photocatalytic potential against NOR can be done in the following points:

Morphology and Photocatalytic Activity

The morphology of CuO nanostructures is a critical factor influencing their photocatalytic efficiency against NOR molecules. The synthesized nanostructures in the present study exhibit a smaller size, a high surface area, a unique shape, and a greater crystalline value, which significantly enhance light absorption and provide a higher density of active sites for the photocatalytic process. These characteristics facilitate efficient electron-hole pair generation and reduce recombination rates, leading to a more effective NOR degradation under visible light irradiation.

Synergistic Effects of CuO Nanostructures

The structural properties of the prepared nanostructures, such as crystallinity, surface defects, and smaller particle size, contribute to their synergistic effects in NOR removal. These features enhance the generation of ROS and improve charge separation, thus boosting the overall photocatalytic activity. Combining these structural factors ensures that the synthesized nanostructures work synergistically to achieve higher degradation efficiency.

Specific Structural Roles

The specific structural roles of the CuO nanostructures, including exposed crystal facets and oxygen vacancies, play a pivotal role in the photocatalytic degradation process. For instance, oxygen vacancies may act as active sites for ROS generation, while exposed facets could facilitate better interaction with NOR molecules, leading to enhanced degradation. These structural roles are integral to the high photocatalytic performance observed in the current study.

2.8.9. Examination of the Turnover Number (TON) and Turnover Frequency (TOF)

The calculation of the Turnover number was done by the formula provided below:

$$\mathrm{T}\mathrm{O}\mathrm{N}={\displaystyle \frac{\mathrm{M}\mathrm{o}\mathrm{l}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{N}\mathrm{O}\mathrm{R} \mathrm{d}\mathrm{e}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{e}\mathrm{d}}{\mathrm{m}\mathrm{o}\mathrm{l}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{c}\mathrm{a}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{y}\mathrm{s}\mathrm{t}}}$$

(3)

To calculate the total moles of NOR degraded by the action of the synthesized photocatalyst, the photocatalytic assay was performed under distinct specific conditions (Exposure to light, pH, time, and dosages of both reactants). To that, the concentration of NOR before and after the degradation was measured by UV-Vis spectroscopy, and the number of moles of NOR degraded by the impact of the photocatalyst during the reaction was calculated. Then, the moles of active sites available were calculated by considering the total amount of photocatalyst as a proxy, and the provided values were included to calculate TON.

On the other hand, TOF was calculated by the formula:

$$\mathrm{T}\mathrm{O}\mathrm{F}={\displaystyle \frac{\mathrm{T}\mathrm{O}\mathrm{N}}{\mathrm{T}\mathrm{i}\mathrm{m}\mathrm{e} \left(\mathrm{s}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{d}\mathrm{s}\right)}}$$

(4)

Upon calculation:

Moles of NOR degraded = 0.001;

Calculated moles of active sites in the synthesized photocatalyst = 0.001;

Overall reaction time taken = 1 h = 3600 s.

Hence,

$$\mathrm{T}\mathrm{O}\mathrm{N}={\displaystyle \frac{0.001}{0.0001}}=10$$

(5)

$$\mathrm{T}\mathrm{O}\mathrm{F}={\displaystyle \frac{10}{3600}}=0.00278 {\mathrm{s}}^{-1}$$

(6)

These values indicate that each active site on the photocatalyst can convert 10 molecules of NOR before deactivation, and the conversion rate is 0.00278 molecules per second per active site.

2.8.10. Mechanism of the ROS Generation Responsible for the Photocatalytic NOR Degradation and Antimicrobial Activity

ROS are crucial in determining photocatalytic and antimicrobial activities. The general mechanism of ROS generation during photocatalysis involves several key steps:

Photon absorption: When the photocatalyst (CuO NPs) is exposed to light, usually in the UV or visible range, it absorbs photons. This energy excites electrons (e⁻) from the valence band to the conduction band, leaving behind positive holes (h⁺) in the valence band.

Formation of ROS: Hydroxyl radicals (^•OH): The photogenerated holes (h⁺) can react with water (H₂O) or hydroxide ions (OH⁻) adsorbed on the photocatalyst surface to produce highly reactive hydroxyl radicals. Superoxide anions (O₂^•−): The excited electrons (e⁻) in the conduction band can reduce oxygen (O₂) molecules present in the environment, generating superoxide anions. These anions can further react with protons (H⁺) to form hydrogen peroxide (H₂O₂), which can then decompose into more hydroxyl radicals. Singlet oxygen (¹O₂): Superoxide anions or other ROS can interact to form singlet oxygen, another reactive species.

Antimicrobial activity: ROS, such as hydroxyl radicals and superoxide anions, are highly reactive and can damage microbial cells by oxidizing vital cellular components, including lipids, proteins, and DNA. This oxidative stress leads to cell membrane disruption, enzyme inhibition, and cell death.

Photocatalytic activity: In photocatalysis, ROS contributes to the NOR degradation. The generated ROS can attack and break down the chemical bonds of pollutants, leading to their mineralization into less harmful compounds, such as CO₂ and H₂O.

2.9. Photocatalytic H₂ Production

The photocatalytic performance of the green-synthesized CuO nanostructures was further extended to determine the H₂ production by photocatalytic reforming. To examine the rates of H₂ production, different doses of photocatalyst (0–10 wt.%) were dispersed in a 150 mL aqueous solution containing 30 vol% ethanol as a sacrificial agent [61]. The results of the H₂ production are presented in Figure 9A–B. It can be seen that we increased the photocatalyst dosage from 0.5 wt.% to 1.5 wt.% of the overall H₂ production tended to increase, reaching 19.52 mmol g⁻¹ h⁻¹. The yield was then compared with commercially available CuO and Pt catalysts to determine the overall cost and H₂ production. As determined by Chen et al., the same dose of Pt/TiO₂ (1.5 wt.%) was 29.5 mmol g⁻¹ h⁻¹, which is somewhat comparable to the engineered photocatalyst and indicates that the prepared photocatalyst has sufficient potential to replace the Pt co-catalysts for H₂ production. Another advantage of the proposed photocatalyst is that there is no need to provide any supplementary H₂ activation step, as in the Pt-constituting photocatalyst, the H₂ reduction is required at 300–400 °C to reduce the Pt species (from Pt²⁺ or Pt⁴⁺ to Pt⁰).

Moreover, different cycles of photocatalytic performance were carried out to evaluate the commercial ability and recyclability of the prepared photocatalyst for H₂ production. The reusability results are presented in Figure 9C. The results indicate that after the first cycle, when the photocatalyst was separated from the system and reused, there was a noticeable decline in H₂ production, with a total of 14.93 mmol g⁻¹ h⁻¹ of H₂ produced. After that, the photocatalyst continued for 5 cycles, and even after five cycles, sufficient hydrogen was produced, as the final cycle yielded 8.85 mmol g⁻¹ h⁻¹ of H₂. This could be due to the reason that biologically synthesized photocatalyst could introduce the formation of structural defects that might be more active in the primary cycle and could be unavailable for further cycles but still sustain marvelous performance. Finally, a comparison table illustrates the differences between the expensive, green-synthesized Pt-containing CuO and the commercially available CuO is provided in Figure 9D. It is important to note that a direct quantitative comparison of H₂ production rates across different studies is complex due to variations in experimental conditions (e.g., light source, intensity, reactor geometry). However, the observed activity of our green-synthesized CuO is notable and demonstrates its potential as a cost-effective, sustainable photocatalyst for H₂ generation, warranting further investigation.

2.10. Antibacterial Activity

The potential of green-synthesized CuO nanostructures was further investigated against antibacterial strains. The antibacterial activity was assessed by measuring the zone of inhibition (in nm) around the wells containing CuO NPs on agar plates inoculated with the bacterial strains. The inhibition zones were measured and compared with those of the controls. Synthesized nanostructures exhibited exceptional antibacterial activity against H1 (Klebsiella) and 38 (Bacillus), as evidenced by readings higher than those of the controls. The results are summarized in

Table 5. Summary of the antibacterial results of the prepared CuO NPs.

Bacterial Strain	DMSO (Control)	Ciprofloxacin (Control)	CuO NPs
H1 (Klebsiella)	3 nm	15 nm	8 nm
-ve (E. coli)	7 nm	10 nm	-
38 (Bacillus)	5 nm	20 nm	9 nm

below. The observed antibacterial activity of CuO NPs suggests that it is particularly effective against Gram-positive bacteria (Bacillus) and Gram-negative bacteria (Klebsiella) but not against E. coli. The differential activity could be attributed to the nanostructures’ unique structural and compositional characteristics, which may interact differently with the cell walls of various bacteria. This is shown in Figure 10A–C.

2.11. In Vitro Anticancer Activity

The synthesized CuO nanostructures were also examined for anticancer activity toward the HeLa cell line, taking doxorubicin as a standard. The results indicated that the CuO nanostructures were found to be comparatively weaker than the standard drug. The value of IC₅₀ of the CuO was found to be 15.3 ± 1.40 μM as compared to the drug (0.9 ± 0.14). On the other hand, the cytotoxic effects of CuO nanostructures were assessed using the MTT assay on the 3T3 cell line. The results demonstrated significant cytotoxicity, with the nanostructures showing an 80.12% inhibition at a 30 µg/mL concentration. The IC₅₀ value was calculated to be 18.04 ± 0.02 µg/mL, suggesting a moderate potency level. Comparatively, the standard anticancer drug Doxorubicin exhibited a higher inhibition rate of 95.8% at the same concentration, with a substantially lower IC₅₀ of 0.17 ± 0.04 µM.

2.12. In Vitro Antifungal Activity

An in vitro antifungal bioassay was conducted using the agar tube dilution protocol to evaluate the antifungal activity of the prepared CuO nanostructures. The test organisms included Trichophyton rubrum, Candida albicans, Aspergillus niger, Microsporum canis, Fusarium lini, Candida glabrata, and Aspergillus fumigatus. The concentration of the silver nanoparticle sample was 3000 µg/mL in 10 µL of DMSO. The linear growth of fungi in the presence of the nanostructures was compared to a control, and the percentage inhibition of fungal growth was calculated. Miconazole and Amphotericin B were used as standard antifungal drugs for comparison. The CuO nanostructures demonstrated significant antifungal activity across all tested fungi, with inhibition rates ranging from 70% to 90%. Detailed results are summarized in

Table 6. Treatment of prepared CuO nanostructures against different fungi.

Fungus	Sample Growth (nm)	Control Growth (nm)	% Inhibition	Standard Drug
Trichophyton rubrum (ATCC MYA 4438)	10	100	90%	Miconazole
Candida albicans (ATCC 36082)	10	100	90%	Amphotericin B
Aspergillus niger (ATCC 1015)	30	100	70%	Miconazole
Microsporum canis (ATCC 10214)	10	100	90%	Miconazole
Fusarium lini (VNRL 2204)	30	100	70%	Miconazole
Candida glabrata (ATCC 2001)	30	100	70%	Miconazole
Aspergillus fumigatus (Clinical isolate)	10	100	90%	Miconazole

.

Experimental setup:

Concentration of Sample: 3000 µg/mL in 10 µL of DMSO;

Incubation Time: 27 h;

Incubation Period: 7–10 days.

2.13. Comparison

Tables S3–S5 provide comparison tables that highlight the advantages and outcomes of the present study across all types of activities.

2.14. Calculation of the Data for Standard Deviation (SD)

The calculation of the SD of the data observed from the reported study was done using the following formula:

$$SD=\surd (\sum (\mathrm{x}-¯\mathrm{x}\left)\right(\mathrm{x}-\mathrm{x}¯)2/\mathrm{n})$$

(7)

Here, X represents individual data points, x̄ is the mean, and n is the total number of data points. Looking at the readings and all of the observed values of photocatalytic degradation (Presented in Table 2), the value of SD was found to be 6.72. On the other hand, the values of H₂ production in the case of each experiment (Presented in Figure 9A,B) displayed the SD values of 0.5 wt.% CuO: 0.154, 1 wt.% CuO: 2.91, 1.5 wt.% CuO: 4.25, 2 wt.% CuO: 6.36, 4 wt.% CuO: 2.89, 6 wt.% CuO: 2.62, 8 wt.% CuO: 1.58, 10 wt.% CuO: 0.97. The SD value of the antibacterial assay, as observed by the action of the prepared nanoparticles, was found to be 0.5 nm. Finally, the standard deviation (SD) for the antifungal inhibition percentages is 9.90%. This indicates the variability in the inhibition results across the different fungal strains tested.

3. Materials and Methods

3.1. Chemicals

Copper nitrate trihydrate (Cu(NO₃) 3H₂O, 99.99%) and ethanol (C₂H₅OH, 99.8%) were procured from Sigma-Aldrich and were used for the experimentation without any further purification. The model pollutant, Norfloxacin (C₁₆H₁₈FN₃O₃, ≥98%), was used for photocatalytic degradation studies, for the biological assays, Nutrient Agar, Ciprofloxacin, Dimethyl Sulfoxide (DMSO), and Normal Saline were used. The bacterial strains tested were Klebsiella sp. (H1), Escherichia coli (-ve), and Bacillus sp. (38). For the anticancer assessment via MTT assay, HeLa and 3T3 cell lines, Minimum Essential Medium Eagle (MEM), Fetal Bovine Serum (FBS), Penicillin–Streptomycin, MTT reagent (3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide), and Doxorubicin were utilized. The antifungal activity was evaluated against Trichophyton rubrum (ATCC MYA 4438), Candida albicans (ATCC 36082), Aspergillus niger (ATCC 1015), Microsporum canis (ATCC 10214), Fusarium lini (VNRL 2204), Candida glabrata (ATCC 2001), and Aspergillus fumigatus (Clinical isolate), with Miconazole and Amphotericin B as standard drugs. The radical scavengers used in the mechanistic study included p-Benzoquinone (p-BQ, C₆H₄(=O)₂, ≥98%), methanol (CH₃OH, ≥99.85%), sodium azide (NaN₃, ≥99.5%), and ethylenediaminetetraacetic acid (EDTA, (HO₂CCH₂)₂NCH₂CH₂N(CH₂CO₂H)₂, 99.4–100.6%). High-purity Nitrogen (N₂) gas was used in the H₂ production test. De-ionized water was used throughout the experiments.

3.2. Preparation of Aqueous Plant Extract

Fresh plant leaves were collected from Bahawalpur city. First, the collected parts were thoroughly washed with tap water and distilled water to remove the dust and contamination. Then, the plant parts were allowed to shade-dry, and once completely dry, they were ground using a mortar and pestle and subjected to extraction. For extraction, the dried plant powder (200 g) was well mixed with 500 mL of de-ionized water in a 1 L glass beaker and stirred for 24 h. After a day of vigorous stirring, the solution was slightly heated up to 45 °C and maintained at this temperature for 30 min to foster a better saturation of phytoconstituents present in the plant crude. After allowing the solution to cool at room temperature, the solution was filtered by Whatman filter paper no.1 (Sigma-Aldrich, Darmstadt, Germany), and the extract was stored at −4 °C for further use.

3.3. Green Synthesis of Copper Oxide Nanostructures

A 0.1 M solution of copper nitrate was prepared in 70 mL of deionized (DI) water. This solution was poured into a conical flask and heated on a hot plate for vigorous stirring. To the precursor solution, 30 mL of Oxystelma esculentum plant extract was added dropwise. After adding plant extract, the solution was heated at 60 °C for 3 h. The solution turned dark brown, indicating the formation of a copper-based precipitate. The entire solution was centrifuged at 6000 rpm for 30 min to separate the precipitate from the solution. After collecting the precipitate, it was washed with ethanol and water to further purify it. It is noteworthy that ethanol was used only for the purification step to ensure clean surfaces and prevent microbial growth during drying, rather than as a reaction solvent. In future optimization, the use of bio-derived or lower-purity ethanol could further enhance the overall greenness of the process. Once the black-coloured powder was obtained, it was calcined at 400 °C for 60 min in air, cooled at 10 °C per minute, and then stored at −4 °C for further experiments.

3.4. Photocatalytic Degradation of NOR by Using Copper Oxide Nanostructures

3.4.1. Central Composite Design (CCD)-Based Optimization Methodology for Photocatalysis

Several experimental methods have been utilized to optimize the numerous parameters. The antibiotic NOR was selected as a model pollutant to estimate the photocatalytic efficacy of the synthesized nanostructures. The design of the experiment (DOE)-based methodologies provide an additional advantage of investigating the combined effects of the understudy variables rather than individual impacts documented for the conventional one-factor-at-a-time (OFAT) approach [62]. The CCD methodology is a statistically based RSM approach where the variables of reaction time (A, min), CuO-NPs dose (B, mg), NOR dose (C, ppm), and pH (D, units) were optimized by utilizing the least-squares techniques. (

Table 7. Summary of the variables utilized for the development of the CCD model.

Variables	Label	Units	Low (−1)	High (+1)	$-\mathit{\alpha }$	$+\mathit{\alpha }$	Mean (0)	Standard Deviation
Reaction time	A	min	30.00	60.00	15.00	75.00	45.00	13.65
CuO-NPs dose	B	mg	20.00	50.00	05.00	65.00	35.00	13.65
NOR dose	C	ppm	20.00	45.00	07.50	57.50	32.50	11.37
pH	D	units	05.00	10.00	02.50	12.50	07.50	02.27

). The details associated with the understudy variables are presented in Table 1. The factorial formula (i.e., a total number of experiments (N) = 2k + 2^k + central points, where k represents the number of the understudy variables) was implemented for the development of the DOE standard table comprising 30 experiments encompassing 06 central, 16 factorial, and 08 axial data points [63]. The CCD-based modeling developed for the photocatalytic reaction of NOR is presented in Equation (1).

$$Photocatalytic efficacy (P\%)={\beta }_o+\sum _{i=1}^k{\beta }_{i}Z+\sum _{i=1}^k{\beta }_{ii}{Z}^2+\sum _{i=1}^{k-1}\sum _{j=2}^k{\beta }_{ij}Z{Z}^{\prime }+ϵ$$

(8)

where $\beta$ represents the working coefficients associated with specific parameters comprising intercept (${\beta }_o$), individual (${\beta }_i$), quadratic (${\beta }_{ii}$), and interaction (${\beta }_{ij}$) terms, with $Z$/$Z^{\prime }$ representing any of the understudy variables under the compulsion that $Z\ne Z^{\prime }$, and the unavoidable error values during the experimentation represented by $ϵ$ [64]. The details of the model’s formation and operation are summarized in the Supplementary Materials.

3.4.2. Experimental Section

The photocatalytic degradation of NOR by CuO-NPs was studied using spectroscopy. The irradiation setup was supplied with a 10-watt mercury lamp (Philips, TUV 48T5 HO 4P SE, manufactured by Signify N.V., Eindhoven, The Netherlands) as the light source, with a primary emission wavelength of 254 nm (UV-C region). The light intensity at the position of the reaction vessel was measured using a UV-light meter (LUTRON UVA-365A, Lutron Electronic Enterprise Co., Ltd., Taipei, Taiwan) and was maintained at 1.5 mW/cm² for all experiments. For the distinctive experiment, the specified amounts of the reaction components (including the CuO-NPs dose and NOR dose) and reaction conditions (including pH and reaction time) as indicated in the generated CCD standard table were implemented. The reaction volume for each particular reaction was set to 50 mL. A 2 mL aliquot of the reaction sample was taken out to acquire the information associated with the concentration/absorption intensity of the NOR at any time ‘t’ (i.e., $A_t$) [65]. The photocatalyst was removed from the reaction medium by centrifugation. The photocatalytic efficacy (P%) of the CuO-NPs was investigated by measuring the percentage degradation of NOR from the formula presented in Equation (2).

$$P\%={\displaystyle \raisebox{1ex}{$A_t-A_o$}\!\left/ \!\raisebox{-1ex}{$A_o$}\right.}\times 100$$

(9)

Here:

$A_t$ and $A_o$ represent the concentration of NOR after treatment at time t and the initial concentration of NOR before treatment [66].

3.4.3. Statistical Analysis

The statistical tests, including the probability (p)-test, Fisher’s (F)-test, coefficient of determination (R²), adjusted-R² test, and predicted-R² test, were performed to assess the validity of the numerous predictive models. The predictive models of linear, quadratic, cubic, and two-factor interaction (2-FI) were tested to examine the validity of the predictive model for representing the acquired results for photocatalytic NOR degradation. The analysis of variance (ANOVA) was also performed to estimate the statistical significance/difference between the understudy variables [67]. Statistical and RSM analyses were performed using Design-Expert (13) software.

3.4.4. CCD-RSM-Based Standard Table Development

Designing any model aims to establish the relationship between the dependent variable (usually called response) and the associated explanatory variables (Z; where Z = A, B, C, D, …) [68,69,70]. In our experiment, the photocatalytic efficacy (P%) of CuO-NPs for NOR degradation was used as the response variable. In contrast, the variables (reaction time (A), CuO-NPs dose (B), NOR (C), and pH (D)) were regarded as the explanatory variables. The development of the CCD was preferred because it is the second-order factorial model capable of performing predictive analysis owing to its rotatability function. The developed CCD model is presented as Equation (3).

$$\widehat{P }\%=\widehat{{\beta }_o}+\widehat{{\beta }_a}A+\widehat{{\beta }_b}B+\widehat{{\beta }_c}C+\widehat{{\beta }_d}D+\widehat{{\beta }_{ab}}AB+\widehat{{\beta }_{ac}}AC+\widehat{{\beta }_{ad}}AD+\widehat{{\beta }_{bc}}BC+\widehat{{\beta }_{bd}}BD+\widehat{{\beta }_{cd}}CD+\widehat{{\beta }_{aa}}{A}^2+\widehat{{\beta }_{bb}}{B}^2+\widehat{{\beta }_{cc}}{C}^2+\widehat{{\beta }_{dd}}{D}^2$$

(10)

where the $\widehat{\beta }$ represents the constant coefficients associated with the intercept (${\beta }_o$), individual factors (${\beta }_a$, ${\beta }_b$, ${\beta }_c$, and ${\beta }_d$), second-order quadratic terms (${\beta }_{aa}$, ${\beta }_{bb}$, ${\beta }_{cc}$, and ${\beta }_{dd}$), and interaction terms (${\beta }_{ab}$, ${\beta }_{ac}$, ${\beta }_{ad}$, ${\beta }_{bc}$, ${\beta }_{bd}$, ${\beta }_{cd}$).

3.5. Photocatalytic H₂ Production Test

The photocatalytic H₂ production was evaluated by water splitting at room temperature in a closed-gas-recirculation system. In a typical experiment, 0.05 g of the photocatalyst was dispersed in a 150 mL aqueous solution containing 30 vol% ethanol, which acts as a sacrificial reagent to consume photogenerated holes and enhance H₂ evolution. The suspension was stirred vigorously. To remove dissolved oxygen, the system was purged with N₂ gas for 30 min. The suspension was then stirred in the dark for an additional 60 min to establish adsorption–desorption equilibrium. The reaction was initiated by irradiating the mixture with a 300 W Xenon lamp (PerfectLight, Beijing, China, PLX-SXE300) equipped with a 400 nm cut-on filter [71]. The light intensity was calibrated to 100 mW/cm² at the reaction vessel surface. The evolved gases were automatically sampled and analyzed at regular intervals using a gas chromatograph (Shimadzu, Nexis™ GC-2030, Kyoto, Japan) equipped with a thermal conductivity detector (TCD), using argon as the carrier gas. The amount of H₂ produced was quantified based on a calibrated standard.

3.6. Antibacterial Assay

The antimicrobial activity of the synthesized nanostructures was evaluated using an agar well diffusion method. The following materials were utilized: Dimethyl Sulfoxide (DMSO), Ciprofloxacin (control), nutrient agar, bacterial strains, Petri plates, test tubes, cotton swabs, normal saline water, and blue tips.

Preparation of nutrient agar plates: 14 g of nutrient agar was dissolved in 500 mL of distilled water and sterilized by autoclaving. The sterilized agar was then poured into 15 Petri plates and allowed to solidify.

Bacterial inoculum preparation: Fresh bacterial cultures were incubated for 24 h at 37 °C. Bacterial colonies were picked using an inoculating loop and mixed with 3 mL of normal saline water in test tubes to prepare the inoculum.

Inoculation and well preparation: Cotton swabs were dipped into the bacterial inoculum and swabbed onto the surface of the nutrient agar plates. Wells were created in the agar plates using a sterile tool and sealed with a drop of agar solution.

Sample and control preparation: The test samples and Ciprofloxacin were mixed in DMSO. Using blue tips, the mixtures were dispensed into the wells on the agar plates.

Incubation: The plates were incubated at 37 °C for 24 h.

Observation: The zone of inhibition was observed and measured the next day.

3.7. Anticancer Activity Protocol

Cell culture and preparation: HeLa cells (cervical cancer) were cultured in Minimum Essential Medium Eagle (MEM), supplemented with 5% fetal bovine serum (FBS), 100 IU/mL penicillin, and 100 µg/mL streptomycin. The cells were maintained in 75 cm² flasks and incubated in a 5% CO₂ atmosphere at 37 °C. Exponentially growing cells were harvested, counted using a hemocytometer, and diluted to a concentration of 6 × 10⁴ cells/mL.

MTT Assay Procedure

Seeding cells: A total of 100 µL of the prepared cell suspension (6 × 10⁴ cells/mL) was seeded into each well of a 96-well flat-bottomed microplate and incubated overnight to allow cell attachment.

Treatment: After incubation, the medium was replaced with 200 µL of fresh medium containing various concentrations of silver nanostructures (1–30 µM).

MTT addition: After 48 h of treatment, 200 µL of MTT solution (0.5 mg/mL) was added to each well, followed by an additional 4 h of incubation.

Formazan solubilization: 100 µL of DMSO was added to each well to solubilize the formazan crystals formed by MTT reduction.

Measurement and analysis: The absorbance of each well was measured at 570 nm using a microplate reader (Spectra Max Plus, Molecular Devices, San Jose, CA, USA). The percentage inhibition of cell growth was calculated using the formula:

$$\% Inhibition=100- \left({\displaystyle \frac{Mean O.D. of test compound-Mean O.D. of-ve control}{Mean O.D. of+ve control-Mean O.D. of-ve control }}\right)\times 100$$

(11)

The IC₅₀ value, representing the concentration of the compound that inhibits 50% of cell growth, was determined from the dose–response curve. Data was processed using SoftMax Pro 6.2 GxP software (Molecular Device).

4. Conclusions

In conclusion, our study introduces a novel methodology that fosters the green synthesis of CuO nanostructures from the aqueous plant extract of Oxystelma esculentum. The study significantly presents and enhances the potential of prepared CuO nanostructures for wastewater treatment, H₂ production, and broad-spectrum biological activities. Through rigorous data-driven methodologies and comparative analyses, we have determined that our approach outperforms prevailing methods in terms of cost-effectiveness, accuracy, and environmental friendliness. The practical implementation of our methodology not only fosters a robust framework for the multi-dimensional potency of green-synthesized nanostructures for biological and environmental applications but also opens new avenues for further development and research. We strongly believe that this progression will significantly contribute to the ongoing efforts to address emerging environmental and pathogenic issues, setting a new standard for future inventions. Future studies could explore the controlled modification of these green-synthesized CuO nanostructures, for instance, through partial reduction to form Cu-CuO heterostructures, to tailor their properties for specific thermocatalytic applications beyond photocatalysis. Future work will also focus on a detailed investigation of the long-term structural stability of the catalyst, particularly the role of the minor Cu₂O phase, through post-reaction characterization to fully elucidate the stability mechanism.