Facet-dependent adsorption and its effect on photocatalytic reactions: insights from diuron degradation on zinc oxide surfaces

Abstract

This study provides direct experimental evidence of facet- and pH-dependent interactions governing the adsorption and photocatalytic degradation of diuron, a persistent herbicide, on zinc oxide (ZnO) surfaces. Unlike conventional studies that assume uniform catalyst surfaces, this work systematically measured the interaction strength on different ZnO facets, using atomic force microscopy (AFM)-based force spectroscopy under conditions replicating the reaction environment. Interaction strengths among ZnO surfaces followed the order: O-terminated > Zn-terminated > mixed-terminated. The finding that ZnO powder, with polar Zn- and O-terminated surfaces, exhibited adsorption capacity approximately five times higher than that of ZnO nanorods, which have non-polar mixed-terminated surfaces, despite having an order of magnitude lower surface area further highlights the role of surface interactions in governing adsorption and subsequent photocatalytic performance. Stronger adsorption under acidic conditions enhanced the degradation efficiency, while alkaline conditions altered adsorption orientations and reduced the capacity. The degradation pathways were found to be highly facet- and pH-dependent, leading to distinct sets of intermediates with varying toxicity. Cytogenotoxicity assays revealed that certain degradation products formed under alkaline conditions were more harmful than diuron itself, underscoring the need for optimized photocatalyst designs. This work provides critical insights into facet-dependent interactions and pH effects, paving the way for more effective and safer photocatalytic solutions for environmental remediation.

---

1. Introduction

Photocatalytic processes have become a promising solution for environmental remediation by degrading pollutants such as dyes, organic compounds, and microorganisms from various industrial sources under light irradiation in an environmentally friendly manner.¹ It is generally accepted that heterogeneous photocatalytic degradation involves the adsorption of target molecules onto the catalyst, the generation of active radicals upon catalyst irradiation, and the oxidation of the adsorbed molecules by the radicals.² Although the photocatalyst surfaces consist of different crystallographic facets, most photocatalysis research treat the catalyst surface as uniform, effectively assuming that the effect of each facet that may interact differently with adsorbed molecules is insignificant relative to the interactions between the adsorbed molecules and the radicals.^3–6 This is due to a lack of experimental data to support an alternative argument, especially from the direct measurements in a liquid environment where photocatalytic reactions take place. This lack of systematically designed experiments further limits the ability to understand and optimize facet-specific adsorption and reaction mechanisms as well as the design of effective photocatalysts.

This critical gap is the main focus of this study. The facet-dependent adsorption and subsequent impact on the photocatalytic degradation of diuron [3-(3,4-dichlorophenyl)-1,1-dimethyl urea], which is a persistent and toxic herbicide widely used in agriculture, on zinc oxide (ZnO) is investigated. ZnO is particularly well-suited for this study as a model system for facet-dependent photocatalysis due to its ability to expose and control various crystallographic facets, including polar (Zn- and O-terminated) and non-polar surfaces, enabling precise investigation of facet-dependent phenomena.⁷ Compared to other widely studied photocatalysts such as TiO₂, ZnO offers several advantages, including enhanced charge carrier mobility,⁸ tunable surface defects (i.e., oxygen vacancies and Zn interstitials), and greater pH-dependent surface charge variability,⁹ which directly influence the adsorption and photocatalytic performance. The choice of diuron for this study is particularly relevant because its degradation reactions could generate intermediates with different levels of toxicity, some of which can be more harmful than the parent compound,¹⁰ underlining the criticality of using catalysts with controlled facets.

Two forms of ZnO are used, namely ZnO powder (with (0001) zinc-terminated and (0001̄) oxygen-terminated as dominating surfaces) and ZnO nanorods (exposing (101̄0) mixed-terminated non-polar surfaces). Previous studies have shown that the surface reactivity of ZnO is facet-dependent, with enhanced adsorption of oxygen-containing species, e.g., OH⁻, O₂, and O₂⁻, on polar surfaces due to the presence of oxygen vacancies.¹¹ Nevertheless, most studies, including our previous work, are either based on inference without direct measurement of the interaction or on the measurement in air, in which the effect of pH could not be included. In this work, atomic force microscopy (AFM) force spectroscopy is performed in aqueous solutions to directly measure interaction on different ZnO surfaces under replicated experimental conditions. Additionally, due to the amphoteric nature of ZnO, surface properties of which are influenced by pH,¹² the interaction strength, the adsorption, as well as the photocatalytic degradation of diuron on different ZnO surfaces in both acidic and alkaline environments are systematically investigated.

Since the specific mechanisms of how surface interactions influence photocatalytic activity are not fully understood, it is usually presumed that effective degradation requires a high surface area, substantial adsorption capacity, and conditions that promote radical generation.^13–15 These factors are often prioritized in catalyst design, with little consideration given to the strength and nature of the interaction between the adsorbed molecule and the catalyst surface. In this study, the interaction strength of ZnO surfaces was systematically evaluated using AFM, along with the adsorption and reaction kinetics measurements. These details would provide a deeper mechanistic insight into the facet-dependent surface interactions and how these interactions vary across different pH conditions as well as their subsequent effects on photocatalytic degradation pathways. The findings suggested that the interaction intensity between adsorbed molecules and specific facets could be more essential to reaction activity than the density of active sites on the surface. The unique interactions on different surfaces of the same catalyst materials also resulted in distinct reaction pathways and various by-products with varying degrees of toxicity. The study underlines the importance of selective facets in catalysis and provides a deeper understanding of heterogeneous photocatalytic degradation.

2. Experimental section

2.1 Preparation of zinc oxide (ZnO)

All chemicals used in the experiments were of analytical reagent grade. Two different surfaces of ZnO were synthesized using distinct techniques as reported in our previous work.¹⁶ In brief, the sol–gel method was employed for ZnO powder. In this method, 3.29 g of zinc acetate was dissolved in 20 ml of ethanol. A mixture of 0.25 ml DI water, 1.58 ml diethanolamine, 5 ml ethanol, and 0.18 ml HCl was gradually added to the zinc acetate solution. The resulting solution was then dried at 80 °C for 24 h and calcined at 500 °C for 3 h to form ZnO powder. ZnO nanorods were prepared through the hydrothermal method by dissolving 1.1 g of zinc acetate in 4 ml of distilled water. Subsequently, 8 M NaOH was added dropwise to the zinc acetate solution. A mixture of 2 ml of this solution, 5 ml of polyethylene glycol, and 20 ml of ethanol was then introduced into a Teflon-lined autoclave. The autoclave was heated to 140 °C for 1 h. After cooling to room temperature, the resulting white precipitate of ZnO nanorods was washed with ethanol and DI water several times before being dried at 60 °C overnight.

2.2 Atomic force microscopy (AFM)-based force spectroscopy

Details on sample preparation for AFM imaging and force spectroscopy are provided elsewhere.¹⁷ Due to the particulate nature of ZnO samples, several preparation steps were optimized to ensure effective ZnO particle immobilization and minimize contamination during imaging. AFM imaging and force spectroscopy were performed using an XE-120 atomic force microscope (Park Systems Corp., South Korea). Two types of AFM probes were employed for the measurement of each tip/sample interaction. Initially, topography images were captured in non-contact mode using a tip (ACTA, Applied NanoStructures, USA) with a nominal radius of approximately 6 nm, mounted on a silicon V-shaped cantilever with a spring constant of 56 N m⁻¹. All images were obtained at controlled room temperature (25 ± 2 °C) and atmospheric pressure, with a scan rate of 0.20 Hz. The scan size was 4 × 4 microns for ZnO particles and clusters and 1 × 1 micron for ZnO nanorods.

After particle positions on the substrate were identified, the probe was switched to a silicon tip (ContAl, Budget Sensors, Innovative Solutions Bulgaria Limited, Sofia, Bulgaria), whose surface would be consistently negatively charged at all pH levels relevant to this study. The AFM probes were calibrated in water over a glass substrate at room temperature using thermal tuning to determine cantilever spring constants. Contact mode imaging was then used to locate the ZnO surface at the predetermined particle positions.

Force curves (force vs. cantilever displacement) were obtained by recording cantilever deflection as the sample surface was moved toward (approach curve) or retracted (retract curve) from the tip. Zero force was defined as the force when the tip and sample were sufficiently separated, with no cantilever deflection being detected. For robust statistical validation, 50 to 70 force curves were collected from various areas/particles on each ZnO particle surface. For ZnO powder, identification of Zn-terminated and O-terminated surfaces was done by comparing the obtained force with those from standard Zn-terminated and standard O-terminated surfaces of a single crystal ZnO wafer (Semiconductor Wafer Inc.).

2.3 Adsorption studies

Aqueous diuron solutions were prepared at concentrations ranging from 0 to 25 mg L⁻¹. The pH of each solution was adjusted using 0.1 M HCl or 0.1 M NaOH. Adsorption experiments were initiated by immersing the catalyst (1 g L⁻¹) in the solution at 25 ± 2 °C, conducted in the absence of light. After 6 hours, which was experimentally determined to be sufficient for reaching adsorption equilibrium, the solution concentration was analyzed using reverse-phase high-performance liquid chromatography (HPLC, Class 10VP, Shimadzu, Japan). Data were collected in triplicate for each concentration across two experiment sets. All chemicals were procured from Sigma-Aldrich.

Adsorption parameters were determined through non-linear regression using OriginPro 2018 (OriginLab, Northampton, MA, USA). The error function was defined to evaluate the isotherm fit to the experimental equilibrium data, with the regression coefficient (R²) and chi-squared statistic selected as key metrics. An R² approaching 1 and a chi-squared value near zero indicated strong model alignment with experimental data.¹⁸

2.4 Photocatalytic degradation

Photocatalytic degradation was conducted in a microreactor. A glass substrate was coated with synthesized ZnO through the spin-coating method. ZnO was stirred in 20 ml of ethanol for 20 minutes, then delicately dropped onto a glass plate rotating under a power supply for 1 minute at a rate of 1 drop per second. Centrifugal force ensured the uniform distribution and coating of the glass plate with the catalyst. These coated glasses were left to dry at 60 °C overnight for ethanol removal. Catalyst adhesion was assessed by passing deionized (DI) water through the coated glass, and the quantity of the released catalyst was quantified using ICP-OES (PerkinElmer, model Optima 7000 DV).

In the assembly of the microreactor, a Teflon sheet with a thickness of 250 micrometers was placed between two glass pieces to define the microchannel's thickness. One glass was coated with the catalyst, and the other, a blank glass, was drilled to facilitate the inlet and outlet streams. All microstructures were assembled using a stainless steel housing.¹⁹ A schematic diagram of the microreactor is provided in the ESI (Fig. S1).† The photocatalytic degradation was operated continuously within the microreactor. The residence time was varied from 1 to 15 minutes by adjusting the flow rate. The initial solution concentration was maintained at 10 ppm, and the pH was adjusted to 4 and 10 using 0.1 M HCl and 0.1 M NaOH, respectively. A 40 W mercury lamp with a spectrum of 350–410 nm served as the light source, positioned at the center for uniform energy distribution. The outlet stream underwent concentration analysis by HPLC (Shimadzu, Class 10VP with a Luna 5μ C18(2) column), while intermediates were identified by LC-MS/MS (Thermo Finnigan, LCQ Advantage).

2.5 Toxicity test

Toxicity can be represented in many ways. Since diuron is used as a herbicide, the cytogenotoxicity test performed on Allium cepa was chosen to present the toxicity of diuron. The dried and oversized roots of A. cepa were removed, and the bulbs were immersed in water for 72 hours in the dark. Approximately roots of 15 mm length were then exposed to the test solution for 3 hours. The root tips in different solutions were cut, placed on a glass slide, fixed in Carnoy's solution for 90 minutes, hydrolyzed with HCl for 10 minutes, washed, and stained with 1% aceto-orcein for 5 minutes. Cell division was observed using a light microscope.²⁰ The mitotic index (MI) and aberration cell-to-dividing cell ratio (ADR) were calculated using eqn (1) and (2), respectively, based on 1000 randomly selected cells for each sample.

3. Results and discussion

3.1 Characterization of ZnO catalysts

ZnO catalysts used in this study exhibit distinct morphological and surface properties critical to understanding their pH-dependent interactions and photocatalytic activity. Detailed characteristics of different surfaces are documented elsewhere,¹⁶ with key points summarized here: both forms of ZnO possess a hexagonal wurtzite crystal structure, confirmed by XRD, with diffraction peaks matching the JCPDS 36-1451 reference. The primary distinction between these catalysts lies in their surface terminations: ZnO nanorods predominantly exhibit nonpolar mixed-terminated surfaces, while ZnO powder features polar Zn- or O-terminated surfaces, as shown by FE-SEM and HR-TEM analyses (Fig. S2†).¹⁶ Nitrogen adsorption isotherms of both catalysts are classified as Type III (IUPAC), indicating their nonporous nature (Fig. S3†). The measured specific surface areas are 1.4 m² g⁻¹ for ZnO powder and 17 m² g⁻¹ for ZnO nanorods, the latter being significantly higher due to morphological differences. XPS analysis shows similar surface compositions for both catalysts, despite ZnO nanorods being uncalcined (Fig. S4†). To better understand pH-dependent surface properties (pH: 4, 7, and 10), XPS measurements were also conducted after soaking the samples in aqueous solutions. This approach contrasts with prior studies that lack aqueous treatment to provide a more realistic assessment of pH-related ZnO surface chemistry changes in solution.

Fig. 1 presents high-resolution XPS spectra of the O 1s peak for both ZnO powder and ZnO nanorods under different pH conditions. Deconvolution of the spectra reveals three peaks: the peak near 530 eV (red line) corresponds to lattice oxygen within ZnO; the peak around 531 eV (purple line) is associated with oxygen vacancies or oxygen-deficient regions, indicating surface alterations by defects; the peak near 532 eV (green line) is attributed to the oxygen in functional groups, particularly hydroxyl groups (OH).²¹ Although these peak positions align with our previous data without aqueous treatment, their relative intensities vary with pH conditions.

Fig. 1 High-resolution XPS O 1s spectra deconvolution of ZnO powder (a1–a3) and ZnO nanorods (b1–b3) under varying pH conditions (pH 4, pH 7, and pH 10).

These variations become more evident when comparing the areas under the curves for oxygen vacancies and hydroxyl groups relative to lattice oxygen, as shown in Table 1. The hydroxyl group content on both ZnO catalysts follows the order: pH 7 > pH 4 > pH 10 > air. (The XPS spectra of ZnO in air were previously published.²²) The intensity of the hydroxyl group of ZnO immersed in pH-adjusted solutions is significantly higher than that of ZnO in air, indicating that ZnO in solution is more readily hydrolyzed to form surface hydroxide layers (Zn–OH), leading to notable changes in surface chemistry compared to pristine ZnO.²³ The hydroxyl formation is more pronounced at neutral pH. The lower intensity of hydroxyl groups at acidic and alkaline pH is likely due to H⁺ or OH⁻ ions inhibiting their formation, reflecting amphoteric surface reactions.²⁴

Table 1 XPS data showing the ratio of the area under the curve of O 1s peaks for ZnO powder and ZnO nanorods obtained under different conditions

Sample name	Conditions	Ovac/Olattice	OOH/Olattice
ZnO powder	In air	0.387	0.171
	pH 4	1.103	0.545
	pH 7	1.144	0.891
	pH 10	0.516	0.385
ZnO nanorods	In air	0.519	0.253
	pH 4	1.588	0.607
	pH 7	1.962	1.060
	pH 10	0.432	0.297

---

The trend in oxygen vacancies aligns with that of surface hydroxyl groups, suggesting a connection between hydroxyl group formation and increased surface reorganization (Table 1). Nevertheless, XPS results alone are not sufficient for a comprehensive understanding of the interaction between adsorbed molecules and the ZnO surfaces. For direct evaluation, atomic force microscopy (AFM) force spectroscopy was used to examine ZnO interaction strengths on each facet under different pH conditions. Integrating XPS with AFM findings will deepen the understanding of pH effects on ZnO surface interactions.

3.2 AFM-based force spectroscopy interpretation

AFM-based force spectroscopy was employed to quantify the ZnO particle adhesion energies under acidic and alkaline conditions, allowing differentiation between Zn- and O-terminated surfaces of ZnO powder. Measurements were conducted in pH-adjusted solutions, with at least 60 force–distance curves per interaction pair. The adhesion energies between silicon probes and ZnO surfaces are summarized in Table 2, with statistical analysis detailed in Table S2.† Notably, the silicon tip surface, naturally oxidized under ambient conditions, forms a silicon oxide layer with silanol groups (Si–OH). Due to its isoelectric point (IEP) of 2–3, the silica surface remains negatively charged across all tested pH values (4 to 9).²⁵

Table 2 Summary of adhesion energies of the interaction between the AFM probe and ZnO surfaces^a

Surfaces	Adhesion energy (×10^−18 J)
	pH 4	pH 7	pH 9
a The presented data are based on averaging results obtained from at least 60 force–distance curves for each interaction pair.
O-Terminated	52.15 ± 4.44	42.67 ± 3.37	9.40 ± 2.47
Zn-Terminated	28.33 ± 4.12	20.90 ± 2.84	31.05 ± 3.69
Mixed-terminated	16.46 ± 3.12	19.16 ± 3.09	23.11 ± 3.36

---

For O-terminated surfaces of ZnO powder, the highest adhesion energy (52.2 × 10⁻¹⁸ J) was observed at pH 4. The magnitude of the adhesion energy indicates a strong polar–polar interaction between the positive charge on O-terminated surfaces and the negatively charged silica tip.¹⁷ The positive charge on the O-terminated surfaces could result from proton uptake that compensates for the charge on surface oxygen.²⁶ As the pH increased, adhesion energy continued to decrease and dropped sharply to 9.4 × 10⁻¹⁸ J at pH 9, primarily due to repulsive electrostatic forces between negatively charged surfaces. This trend reflects the weaker attraction at higher pH values, as protonation on the O-terminated surfaces diminishes.

For Zn-terminated surfaces of ZnO powder, the adhesion energy to negatively charged silica tips decreased from 28.3 × 10⁻¹⁸ J at pH 4 to 20.9 × 10⁻¹⁸ J at pH 7 and subsequently increased to 31 × 10⁻¹⁸ J at pH 9. This pattern correlates with XPS results of hydroxyl group intensity on the surface, which follows the order pH 7 > pH 4 > pH 10, suggesting a strong influence of the hydroxyl group on the interaction between Zn-terminated surfaces and the silica tips. The partial positive charges on hydrogen atoms within the hydroxyl groups could enhance electrostatic attraction with a negatively charged AFM tip. However, the increased hydroxyl density would increase the number of negatively charged O atoms that could impede the effective rearrangement of hydrogen atoms, affecting adhesion energy trends on Zn-terminated surfaces (Fig. 2).

Fig. 2 Comparison of adhesion energies between the AFM probe and ZnO surfaces under different pH conditions.

In the case of mixed-terminated ZnO nanorods, the adhesion energy was relatively low and changed less dramatically across pH levels compared to both Zn- and O-terminated surfaces of ZnO powder. This observation is consistent with previous findings that nonpolar wurtzite surfaces exhibit stable surfaces and low interaction strength.¹⁷ The adhesion energy increased from 16.5 × 10⁻¹⁸ J at pH 4 to 23.1 × 10⁻¹⁸ J at pH 9, with more pronounced increases under alkaline conditions. It should be noted that the interatomic Zn–O distance on the mixed-terminated surface is about 0.2 nm,²⁷ while AFM tips are approximately 10 nm, leading to a distributed interaction across the surface rather than with specific Zn or O atoms. Therefore, unlike the polar Zn- and O-terminated surfaces, the mixed-terminated nanorods show more stable charge distribution and less pronounced pH-dependent variation in adhesion energy.

3.3 Diuron adsorption

The adsorption process primarily depends on the surface properties of the adsorbent and strongly influences subsequent reactions such as photodegradation. Previous characterization of ZnO surfaces revealed pH-dependent variation in surface properties and interaction strengths, which are expected to impact diuron adsorption behavior. This section thus examines the effects of these properties on diuron adsorption onto ZnO surfaces. The Freundlich model was applied to analyze the diuron adsorption isotherms on ZnO catalysts, as it effectively accounts for surface heterogeneity. This model, based on monolayer adsorption on a heterogeneous surface, uses two parameters: K_F for binding capacity and n_F for surface heterogeneity.²⁸ In contrast to the Langmuir model, which assumes a homogeneous surface with identical adsorption sites,²⁹ the Freundlich model provides a better fit and is consistent with the AFM results, which indicate a variation of adsorption energies on ZnO surfaces (see the Langmuir model fitting in Fig. S5,† with the corresponding parameters summarized in Table S1†).

Fig. 3 illustrates the diuron adsorption isotherms for ZnO samples at different pH levels (4, 7, and 10), with nonlinear regression analysis data summarized in Table 3. The model fits the experimental data well, as indicated by low chi-squared values and high R² values. ZnO powder shows a higher heterogeneity parameter, n_F, than ZnO nanorods, indicating greater surface heterogeneity, likely due to the presence of both O- and Zn-terminated surfaces. Conversely, the mixed-terminated surfaces of ZnO nanorods exhibit a more uniform distribution of adsorption energies. Additionally, n_F values for ZnO nanorods decrease below 1 as the pH increases, indicating less favorable adsorption. In contrast, the n_F values for ZnO powder remain stable to pH. This consistency may result from balanced interactions on O- and Zn-terminated surfaces, with relatively constant adsorption energy distribution across pH levels. AFM results in Fig. 2 support this stability, showing consistent differences between the adhesion energies of Zn- and O-terminated surfaces across the pH range.

Fig. 3 Freundlich adsorption isotherms of diuron on ZnO catalysts at different pH levels.

Table 3 Fitted parameters of the Freundlich adsorption model for diuron adsorption on ZnO catalysts

Sample	pH	Freundlich parameter		R ^2	Chi-squared
		n F	K F
ZnO powder	4	1.11	0.149	0.9990	0.0045
	7	1.12	0.093	0.9959	0.0084
	10	1.12	0.071	0.9980	0.0020
ZnO nanorods	4	1.07	0.032	0.9988	0.0009
	7	0.99	0.018	0.9949	0.0025
	10	0.95	0.012	0.9750	0.0112

---

In Fig. 3, the adsorption isotherms of ZnO powder at low diuron concentrations (0–25 mg L⁻¹) are above those of ZnO nanorods across all pH levels, with a steeper slope and a higher binding capacity parameter, K_F (Table 3). This suggests that ZnO powder has a higher binding capacity for diuron despite its smaller surface area (1.4 m² g⁻¹vs. 17 m² g⁻¹ for nanorods), indicating that adsorption capacity is more influenced by surface interactions than the surface area alone. These findings align with the suggestion by prior research that the polar ZnO surfaces interact more strongly with adsorbed species compared to non-polar ZnO surfaces.^30,31 The direct measurement via AFM also shows higher adhesion energies for ZnO powder than ZnO nanorods across the pH range, implying more favorable surface characteristics for diuron adsorption. The stronger surface interactions enhance the adsorption capacity.

Fig. 3 also illustrates the effect of pH on the diuron adsorption capacity for ZnO samples. Both ZnO powder and nanorods exhibit reduced diuron adsorption at higher pH, aligning with decreases in K_F. This indicates more efficient adsorption under acidic conditions. The ZnO powder and nanorods have points of zero charge (PZC) at pH 7.9 and 6.7, respectively, implying that they are positively charged below these pH values and negatively charged above them. This observation challenges the view that the overall electrostatic charge of a particle determines how the adsorption capacity changes with pH,³² as diuron (pK_a of 13.5) remains neutral at most environmental pH levels. Instead, the local interaction between functional groups of diuron and the specific surface of ZnO should be considered. Under acidic conditions (pH 4), diuron adsorption onto ZnO is driven primarily by strong polar–polar interactions between its negatively polar regions (e.g., oxygen and chlorine atoms) and the positively charged ZnO surfaces. This is corroborated by AFM results, which show strong polar–polar interactions under acidic conditions. At alkaline pH (pH 10), diuron adsorption is driven by interactions between the negatively charged ZnO surfaces and the locally positively charged regions of diuron, such as the aliphatic methyl group, resulting in weaker binding and lower adsorption capacity.

Overall, ZnO powder shows a higher adsorption capacity than ZnO nanorods across all studied pH levels, despite having an order of magnitude lower surface area, due to stronger interactions with diuron. Additionally, the pH markedly influences the nature of these surface interactions, affecting diuron adsorption behavior in distinct ways.

3.4 Photocatalytic reaction of diuron

The photocatalytic degradation mechanism involves exposing ZnO to UV light, which provides a larger photon energy than the optical bandgap of ZnO (around 3.1 eV for both ZnO nanorods and powder, as measured using reflective UV/visible spectroscopy) and generates electron–hole pairs upon excitation. The holes react with OH⁻ ions or water to produce hydroxyl radicals (˙OH), as the valence band's potential of ZnO exceeds the redox potentials of ˙OH/OH⁻ (1.99 eV) and ˙OH/H₂O (2.27 eV).³³ Alkaline conditions typically enhance ˙OH formation,¹⁵ as the increased concentration of OH⁻ ions can directly react with the holes in the ZnO's valence band. Although many kinds of active species have been reported to be involved in photocatalytic reactions,³⁴ hydroxyl radicals are widely recognized as the primary contributors to the degradation of pollutants due to their strong oxidation potential and high reactivity with organic compounds.^35–37 Analysis of the molecular structure of the degradation intermediates could also provide confirmation of the participation of hydroxyl radicals in the reaction.

As noted previously, surface interactions at different pH levels significantly influenced diuron adsorption characteristics on ZnO surfaces, which include the adsorption capacity and adsorption orientation, and this adsorption behavior appears to influence how diuron interacts with the radicals formed on the surfaces to form degradation intermediates. The following sections explore its impact on diuron degradation kinetics and the associated reaction pathways.

3.4.1 Diuron degradation kinetics. The experimental data for various residence times align well with a commonly applied kinetics model in photocatalysis,^38–41i.e., the pseudo-first-order model, in which the degradation rate is directly proportional to the remaining diuron concentration. Fig. 4 shows the photocatalytic degradation kinetics of diuron under varying pH conditions with ZnO powder and nanorods. Both forms of ZnO exhibit similar degradation behavior under the same pH conditions, achieving approximately 85%, 82%, and 75% diuron removal at pH 4, pH 7, and pH 10, respectively, after 15 minutes. This trend suggests that diuron degradation is more favorable under acidic conditions. Furthermore, Table 4 shows a notable decrease in the apparent rate constant (k) as the pH increases for both ZnO samples.

Fig. 4 Comparison of photocatalytic degradation kinetics of diuron on ZnO powder and ZnO nanorods under different pH conditions. The error bars are intentionally shifted to the side for clarity of the plot.

Table 4 Degradation kinetic parameters of diuron in photocatalytic degradation using ZnO powder and ZnO nanorods

Sample	pH	k (min^−1)	R ^2	Chi-squared
Powder	4	0.173	0.950	0.122
	7	0.153	0.953	0.091
	10	0.133	0.920	0.118
Nanorods	4	0.183	0.941	0.121
	7	0.154	0.947	0.082
	10	0.119	0.954	0.064

---

Interestingly, despite enhanced OH production under alkaline conditions, diuron degradation does not increase with pH, indicating that hydroxyl radical availability is not rate-limiting and is already adequate in this system. Consistently, the degradation rate dependency on the diuron concentration, according to the pseudo-first-order kinetics model, suggests that photocatalytic efficiency in this system is primarily driven by adsorption characteristics rather than the availability of radicals. Despite ZnO nanorods having a larger surface area than ZnO powder, the degradation rates of diuron between the two catalysts exhibit no significant differences. This finding supports the literature, indicating that photocatalytic performance is not exclusively dictated by the surface area.⁴² Instead, the primary driver of diuron degradation is the interaction between ZnO surfaces and diuron molecules. Under acidic conditions, stronger interactions enhance diuron adsorption, bringing molecules closer to the surface and increasing the likelihood of degradation. These results highlight the critical role of surface interactions in photocatalytic activity.

3.4.2 Reaction pathways. Degradation intermediates could be distinctly detected when the residence time in the microreactor was at least 1 minute; hence, it is inferred that the intermediates detected at this initial residence time resulted from the direct attack of the radicals on the adsorbed diuron. As the residence time increased, these intermediates degraded further until complete mineralization. The molecular structures of these intermediates are detailed in Table S3.† The degradation pathways were inferred from the molecular structures of the intermediates. It should be noted that almost all intermediates that directly descended from diuron show the presence of a hydroxyl functional group; hence, it is consistent with our assumption that the hydroxyl radicals are the primary contributors to diuron degradation.

Facet-dependent pathways for the photocatalytic degradation of diuron on ZnO at neutral pH were introduced in our previous work.¹⁶ In this work, the influence of various pH was studied. LC-MS/MS analysis of diuron photocatalytic degradation intermediates on ZnO samples at pH 4 is shown in Fig. 5. The majority of intermediates detected under acidic conditions were not formed at neutral pH, confirming that substantial change in diuron/ZnO surface interactions due to pH change drastically affects diuron degradation. In the initial stage, the hydroxyl radicals attack both the aliphatic and aromatic regions of diuron on both ZnO catalysts, indicating that diuron adsorbs on the ZnO surface in a manner that makes both regions accessible. This accessibility aligns with the strong interactions between the negatively charged regions of diuron (e.g., chlorine on the aromatic ring and oxygen on the aliphatic side) and the positively charged ZnO surfaces, as indicated by AFM results. Under acidic conditions, the order of interaction strength among ZnO surfaces is ranked as follows: O-terminated, Zn-terminated, and mixed-terminated surfaces, respectively. These interactions likely enhance diuron's accessibility, facilitating hydroxyl radical attack across the molecule.

Fig. 5 Diuron degradation intermediates identified by LC-MS/MS and the proposed degradation pathway at pH 4. Intermediates detected with ZnO nanorods and ZnO powder are shown in red and blue, respectively, while those common to both catalysts are shown in purple.

At pH 10, the degradation pattern changes almost completely, especially for the degradation on ZnO powder. The pH affects polar surfaces more strongly, as witnessed by AFM results, consequently altering the adsorption posture of the diuron. For ZnO nanorods, radicals initially attack only the aliphatic side of the diuron, while on ZnO powder, both aliphatic and aromatic regions are targeted (Fig. 6). On ZnO powder, negatively charged hydroxide ions predominantly adsorb onto Zn-terminated surfaces, attracting the positively charged methyl groups from the aliphatic side. In contrast, diuron's aromatic region likely interacts with O-terminated surfaces on ZnO powder, which electrostatically repels the oxygen and chlorine atoms. The interaction potentially causes a reorientation of hydrogen atoms of the methyl group and the aromatic ring toward the surface and enables hydroxyl attack on both sides. The formation of identical intermediates (C3 and C18) from the aliphatic regions on both ZnO samples suggests similar adsorption characteristics on Zn-terminated surfaces of ZnO powder and mixed-terminated surfaces of ZnO nanorods.

Fig. 6 Diuron degradation intermediates identified by LC-MS/MS and the proposed degradation pathway at pH 10. Intermediates detected with ZnO nanorods and ZnO powder are shown in red and blue, respectively, while those common to both catalysts are shown in purple.

These findings highlight the role of ZnO surface interactions in shaping the degradation pathway and specific products of diuron.

3.5 Proposed diuron adsorption mechanisms

The pH-dependent adsorption behavior of diuron on ZnO surfaces is schematically illustrated in Fig. 7. Under acidic conditions (pH 4), the O-terminated surfaces undergo protonation, resulting in a net positive charge (Fig. 7a). This charge attracts the oxygen of diuron's carbonyl group and the chlorine in the aromatic ring while repelling the methyl groups attached to the nitrogen. On the Zn-terminated surfaces, the partially positive hydroxyl groups also interact with the diuron's negatively charged sites, i.e., chlorine and oxygen (Fig. 7c). However, the presence of hydroxyl groups weakens electrostatic interactions on Zn-terminated surfaces compared to positively charged O-terminated surfaces. This result is supported by AFM measurements, showing higher adhesion energy on the O-terminated surfaces than on the Zn-terminated surfaces in an acidic environment.

Fig. 7 Schematic representation of pH-dependent diuron adsorption mechanisms on ZnO surfaces: (a and b) O-terminated surfaces under acidic and alkaline conditions; (c and d) Zn-terminated surfaces under acidic and alkaline conditions; (e and f) mixed-terminated ZnO nanorod surfaces under acidic and alkaline conditions, respectively.

In contrast, under alkaline conditions (pH 10), O-terminated surfaces become electron-rich, creating repulsive electrostatic forces against the chlorine and oxygen atoms of diuron. This repulsion could lead to the reorientation of the diuron's positively charged hydrogen atoms toward the surface (refer to Fig. 7b). Meanwhile, hydroxide ions on the Zn-terminated surfaces create a negative environment that repels the aromatic end of diuron, leaving only the methyl group hydrogens to adsorb onto the surface through weak van der Waals interactions (Fig. 7d). This disordered adsorption configuration under alkaline conditions reduces the accessible adsorption sites, as reflected in the binding capacity (K_F), which decreases by approximately 37% at pH 7 and 52% at pH 10 compared to that at pH 4.

For ZnO nanorods (Fig. 7e and f), the adsorption behavior mirrors that of ZnO powder, with lower adsorption capacity at higher pH levels. Diuron adsorption on the mixed-terminated surfaces of ZnO nanorods under acidic conditions (Fig. 7e) is comparable to that on O- and Zn-terminated surfaces under alkaline conditions (Fig. 7f) but shows lower capacity across pH levels. This reduced capacity is attributed to the reduced polarity and attractiveness of mixed-terminated surfaces.

3.6 Toxicity assessment

Discussions in previous sections have highlighted the effects of facet-dependent interactions on adsorption, reaction activity, and reaction pathways. It may not be critical for applications like photocatalytic dye degradation or water splitting, as the focus is primarily on achieving high conversion. However, for the degradation of toxic compounds, these effects become particularly important because of the need for precise control toward low-toxicity pathways.

This work employs cytogenotoxicity, indicated by MI and ADR values calculated using eqn (1) and (2), respectively, to represent the toxicity of diuron. Large MI refers to flourishing cell division, indicating low toxicity. On the other hand, high ADR reflects abnormalities in various phases of cell mitosis, hence high toxicity. Low MI and high ADR from diuron solution, compared to the values from deionized water (see Table 5), underline the inherent cytogenotoxicity of diuron. It is also observed that both acidity and basicity also introduce chromosomal irregularities to the cells. Cytogenotoxicity testing was selected among various toxicity assessments due to its relevance to the cellular toxification process in plant cells induced by herbicides such as diuron. It evaluates chromosomal integrity and mitotic activity, serving as an early indicator of genotoxic effects. The cytogenotoxicity demonstrated herein exemplifies how the reaction pathway changes according to the interaction between the molecules and specific facets of the catalyst. Alternative forms of poisoning may produce varying outcomes.

Table 5 Mitotic indexes and chromosomal aberration for mitosis of Allium cepa exposed to the degradation products

Treatment	MI (%)		ADR (%)
	On ZnO powder	On ZnO nanorods	On ZnO powder	On ZnO nanorods
At pH 4
Deionized water	14.45 ± 0.92		14.19 ± 2.38
10 ppm diuron solution	7.45 ± 1.63		29.53 ± 0.77
Diuron solution diluted to the concentration corresponding to 1 min degradation	8.95 ± 2.62	9.00 ± 1.84	25.14 ± 5.20	22.22 ± 1.43
1 min degradation	6.55 ± 0.78	7.60 ± 0.28	38.17 ± 6.74	28.95 ± 2.94
15 min degradation	10.40 ± 0.28	11.05 ± 0.49	20.67 ± 1.48	16.74 ± 1.17
At pH 7
Deionized water	19.70 ± 0.85		7.87 ± 0.74
10 ppm diuron solution	9.45 ± 1.77		21.69 ± 1.85
Diuron solution diluted to the concentration corresponding to 1 min degradation	13.50 ± 1.70	13.10 ± 0.99	12.22 ± 1.48	16.79 ± 1.27
1 min degradation	10.30 ± 0.14	11.45 ± 1.91	21.84 ± 1.76	20.09 ± 5.25
15 min degradation	15.15 ± 0.92	18.90 ± 1.84	12.54 ± 0.18	8.99 ± 2.38
At pH 10
Deionized water	12.55 ± 3.04		16.33 ± 2.31
10 ppm diuron solution	6.15 ± 0.92		41.46 ± 7.43
Diuron solution diluted to the concentration corresponding to 1 min degradation	8.80 ± 0.85	9.00 ± 0.78	26.70 ± 9.74	20.00 ± 9.69
1 min degradation	6.95 ± 0.49	8.10 ± 0.85	38.13 ± 3.74	24.07 ± 3.42
15 min degradation	11.30 ± 0.64	6.85 ± 0.64	19.03 ± 0.81	25.55 ± 7.57

---

In general terms, the degradation products using a 1-minute residence time exhibited increased toxicity compared with the reference solution with the same concentration of diuron. This indicated that certain intermediates from direct diuron degradation become more toxic than diuron itself, especially those formed on ZnO powder. Therefore, considering the disappearance of diuron alone is insufficient for the remediation. Extending the degradation time to 15 minutes revealed a substantial reduction in toxicity because of the continuous degradation of both diuron and its degradation intermediates toward more benign compounds. Nevertheless, the crucial impact of the facet-dependent pathway can be clearly observed herein. Although the photocatalytic degradation of diuron on ZnO nanorods generally yields lower toxicity than that on ZnO powder, the product from 15-minute degradation on ZnO nanorods at pH 10 retained toxicity equivalent to the initial diuron solution, even after roughly 75% of diuron had already been degraded. The results not only emphasize the importance of facet-dependent adsorption and its consequence on reaction pathways but also the effect of pH, which could completely alter the adsorption. The design of the effective photocatalyst for proper remediation must incorporate all aspects affecting interactions between the molecules being degraded and the surface of the catalyst.

Overall, the degradation efficiency increased as the pH decreased, with pH 4 yielding the highest photocatalytic degradation efficiency. The fact that pH 10 offered the lowest degradation efficiency despite the enhanced formation of hydroxyl radicals under alkaline conditions and the larger surface area nanorods exhibited comparable degradation efficiency to ZnO powder emphasize the dominant role of adsorption characteristics and surface interactions in degradation kinetics. Toxicity assessment revealed that ZnO nanorods generally yielded lower toxicity in degradation products, except at pH 10, where cytogenotoxicity persisted despite the high degradation efficiency. These findings highlight the importance of surface interactions, which vary according to pH, in designing ZnO-based photocatalysts for effective remediation. Both the degradation efficiency and potential toxicity of the reaction intermediates must be considered in the design process.

4. Conclusion

This work provides direct experimental evidence that the adsorption strength and photocatalytic degradation efficiency of diuron on ZnO are strongly influenced by facet-dependent interactions, which are further modulated by the pH of the solution. Using atomic force microscopy (AFM)-based force spectroscopy under conditions replicating the reaction environment, we demonstrated that Zn-terminated and O-terminated surfaces exhibit significantly stronger interactions than the non-polar mixed-terminated surface. In detail, the attraction between functional groups in the adsorbed molecule and charges on the catalyst surface, which are markedly affected by the pH of the solution, plays a significant role in adsorption. The adsorption capacity of ZnO powder with polar surfaces was over five times higher than that of ZnO nanorods with non-polar surfaces across all tested pH levels, highlighting the dominant role of surface interaction strength over the surface area alone. These factors subsequently affect the reaction kinetics and reaction pathways in a more profound manner than traditionally assumed.

For the photocatalytic degradation of diuron, its degradation intermediates and associated toxicity vary significantly across different facets of ZnO and under different pH conditions. ZnO powder exhibited superior degradation performance despite its smaller surface area. The degradation of diuron followed pseudo-first-order kinetics, and the presence of hydroxyl functional groups in almost all detected intermediates suggests that hydroxyl radicals are the primary reactive species responsible for the degradation. Acidic conditions facilitated stronger polar interactions and higher degradation efficiency, while alkaline conditions altered adsorption orientations and reduced the efficiency regardless of the fact that the generation of hydroxyl radicals was enhanced under alkaline conditions. Cytogenotoxicity testing also revealed that certain degradation products formed under alkaline conditions were more harmful than diuron itself.

These findings emphasize that designing an effective photocatalyst for optimal remediation requires considering all factors influencing the interactions between the degrading molecules and the catalyst surface. Furthermore, achieving complete degradation of both the target molecules and their intermediates is essential to mitigating environmental risks. This study underscores the importance of facet-specific considerations and pH effects in catalyst design, providing valuable insights for the development of tailored photocatalytic systems for environmental remediation.

Data availability

The data supporting this article have been included as part of the ESI.†

Author contributions

Panuwat Lawtae: conceptualization, data curation, validation, formal analysis, visualization, writing – original draft, writing – review & editing. Sutaporn Meephon: investigation, methodology. Vipada Dokmai: investigation, methodology. Rungthiwa Methaapanon: conceptualization, formal analysis, writing – review & editing. Varong Pavarajarn: conceptualization, validation, formal analysis, writing – review & editing, supervision. All authors have approved the final version of the article.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

This research project was supported by the Second Century Fund (C2F), Chulalongkorn University for the Postdoctoral Fellowship (to Panuwat Lawtae). The authors also acknowledge Naruemon Chumjai and Alisa S. Vangnai, Biocatalyst and Environmental Biotechnology Research Unit, Department of Biochemistry, Chulalongkorn University, for assistance with toxicity assessments.

References

1. D. Lahiri, M. Nag, and R. R. Ray, 17 – Industrial applications of photocatalytic methods such as textile pharmaceutical industries, tannery, and craft, in Photocatalytic Degradation of Dyes, ed. M. Shah, S. Dave, and J. Das, Elsevier, 2021, pp. 467–488.
2. J. Schneider, M. Matsuoka, M. Takeuchi, J. Zhang, Y. Horiuchi, M. Anpo and D. W. Bahnemann, Understanding TiO2 Photocatalysis: Mechanisms and Materials, Chem. Rev., 2014, 114(19), 9919–9986.
3. N. Verma, S. Bhatia and R. K. Bedi, Role of pH on electrical, optical and photocatalytic properties of ZnO based nanoparticles, J. Mater. Sci.: Mater. Electron., 2017, 28(13), 9788–9797.
4. V. V. Gawade, S. R. Sabale, R. S. Dhabbe, S. V. Kite and K. M. Garadkar, Bio-mediated synthesis of ZnO nanostructures for efficient photodegradation of methyl orange and methylene blue, J. Mater. Sci.: Mater. Electron., 2021, 32(24), 28573–28586.
5. J. Q. Lopes and B. C. B. Salgado, Photocatalytic performance of titanium and zinc oxides in diuron degradation, Environ. Challenges, 2021, 4, 100186.
6. M. Yeganeh, E. Charkhloo, H. R. Sobhi, A. Esrafili and M. Gholami, Photocatalytic processes associated with degradation of pesticides in aqueous solutions: Systematic review and meta-analysis, Chem. Eng. J., 2022, 428, 130081.
7. C. Wöll, The chemistry and physics of zinc oxide surfaces, Prog. Surf. Sci., 2007, 82(2), 55–120.
8. S. Hernández, D. Hidalgo, A. Sacco, A. Chiodoni, A. Lamberti, V. Cauda, E. Tresso and G. Saracco, Comparison of photocatalytic and transport properties of TiO2 and ZnO nanostructures for solar-driven water splitting, Phys. Chem. Chem. Phys., 2015, 17(12), 7775–7786.
9. Y. Sun, W. Zhang, Q. Li, H. Liu and X. Wang, Preparations and applications of zinc oxide based photocatalytic materials, Adv. Sens. Energy Mater., 2023, 2(3), 100069.
10. S. Giacomazzi and N. Cochet, Environmental impact of diuron transformation: a review, Chemosphere, 2004, 56(11), 1021–1032.
11. A. McLaren, T. Valdes-Solis, G. Li and S. C. Tsang, Shape and Size Effects of ZnO Nanocrystals on Photocatalytic Activity, J. Am. Chem. Soc., 2009, 131(35), 12540–12541.
12. I. Kazeminezhad and A. Sadollahkhani, Influence of pH on the photocatalytic activity of ZnO nanoparticles, J. Mater. Sci.: Mater. Electron., 2016, 27(5), 4206–4215.
13. K. S. Ranjith and R. T. Rajendra Kumar, Regeneration of an efficient, solar active hierarchical ZnO flower photocatalyst for repeatable usage: controlled desorption of poisoned species from active catalytic sites, RSC Adv., 2017, 7(9), 4983–4992.
14. X. He, Y. Yang, Y. Li, J. Chen, S. Yang, R. Liu and Z. Xu, Effects of structure and surface properties on the performance of ZnO towards photocatalytic degradation of methylene blue, Appl. Surf. Sci., 2022, 599, 153898.
15. Q. Yang, T.-S. Lin, C. Burton, S.-H. Park and Y. Ma, Physicochemical insights of irradiation-enhanced hydroxyl radical generation from ZnO nanoparticles, Toxicol. Res., 2016, 5(2), 482–491.
16. S. Meephon, T. Rungrotmongkol, S. Puttamat, S. Praserthdam and V. Pavarajarn, Heterogeneous photocatalytic degradation of diuron on zinc oxide: Influence of surface-dependent adsorption on kinetics, degradation pathway, and toxicity of intermediates, J. Environ. Sci., 2019, 84, 97–111.
17. V. Dokmai, W. Kundhikanjana, N. Chanlek, K. Sinthiptharakoon, U. Sae-Ueng, W. Phuthong and V. Pavarajarn, Effects of catalyst surfaces on adsorption revealed by atomic force microscope force spectroscopy: photocatalytic degradation of diuron over zinc oxide, Phys. Chem. Chem. Phys., 2020, 22(26), 15035–15047.
18. K. Y. Foo and B. H. Hameed, Insights into the modeling of adsorption isotherm systems, Chem. Eng. J., 2010, 156(1), 2–10.
19. W. Khongthon, G. Jovanovic, A. Yokochi, P. Sangvanich and V. Pavarajarn, Degradation of diuron via an electrochemical advanced oxidation process in a microscale-based reactor, Chem. Eng. J., 2016, 292, 298–307.
20. M. K. Sipahutar and A. S. Vangnai, Role of plant growth-promoting Ochrobactrum sp. MC22 on triclocarban degradation and toxicity mitigation to legume plants, J. Hazard. Mater., 2017, 329, 38–48.
21. T. J. Frankcombe and Y. Liu, Interpretation of Oxygen 1s X-ray Photoelectron Spectroscopy of ZnO, Chem. Mater., 2023, 35(14), 5468–5474.
22. S. Meephon, T. Rungrotmongkol, N. Kaiyawet, S. Puttamat and V. Pavarajarn, Surface-Dependence of Adsorption and Its Influence on Heterogeneous Photocatalytic Reaction: A Case of Photocatalytic Degradation of Linuron on Zinc Oxide, Catal. Lett., 2018, 148(3), 873–881.
23. M. Schindler, F. C. Hawthorne, M. S. Freund and P. C. Burns, XPS spectra of uranyl minerals and synthetic uranyl compounds. II: The O 1s spectrum, Geochim. Cosmochim. Acta, 2009, 73(9), 2488–2509.
24. A. Degen and M. Kosec, Effect of pH and impurities on the surface charge of zinc oxide in aqueous solution, J. Eur. Ceram. Soc., 2000, 20(6), 667–673.
25. I. Sokolov, Q. K. Ong, H. Shodiev, N. Chechik, D. James and M. Oliver, AFM study of forces between silica, silicon nitride and polyurethane pads, J. Colloid Interface Sci., 2006, 300(2), 475–481.
26. S. Su, I. Siretanu, D. van den Ende, B. Mei, G. Mul and F. Mugele, Facet-Dependent Surface Charge and Hydration of Semiconducting Nanoparticles at Variable pH, Adv. Mater., 2021, 33(52), 2106229.
27. V. A. Ermoshin and V. A. Veryazov, Electronic Structure Investigation of Bulk ZnO and Its (1010) Surface, Phys. Status Solidi B, 1995, 189(2), K49–K53.
28. H. M. F. Frenudlich, Over the adsorption in solution, J. Phys. Chem., 1906, 57, 385–471.
29. I. Langmuir, The Adsorption Of Gases On Plane Surfaces Of Glass, Mica And Platinum, J. Am. Chem. Soc., 1918, 40(9), 1361–1403.
30. P. Gorai, E. G. Seebauer and E. Ertekin, Mechanism and energetics of O and O2 adsorption on polar and non-polar ZnO surfaces, J. Chem. Phys., 2016, 144(18), 184708.
31. J. Yang, J. Wang, X. Li, J. Lang, F. Liu, L. Yang, H. Zhai, M. Gao and X. Zhao, Effect of polar and non-polar surfaces of ZnO nanostructures on photocatalytic properties, J. Alloys Compd., 2012, 528, 28–33.
32. T. S. Anirudhan, F. Shainy and A. Manasa Mohan, Fabrication of zinc oxide nanorod incorporated carboxylic graphene/polyaniline composite and its photocatalytic activity for the effective degradation of diuron from aqueous solutions, Sol. Energy, 2018, 171, 534–546.
33. A. D. Lopis, K. S. Choudhari, R. Sai, K. S. Kanakikodi, S. P. Maradur, S. A. Shivashankar and S. D. Kulkarni, Laddered type-1 heterojunction: Harvesting full-solar-spectrum in scavenger free photocatalysis, Sol. Energy, 2022, 240, 57–68.
34. A. A. Mashentseva, D. T. Nurpeisova and M. Barsbay, Effect of copper doping on the photocatalytic performance of Ni2O3@PC membrane composites in norfloxacin degradation, RSC Adv., 2024, 14(7), 4424–4435.
35. Y. Nosaka and A. Nosaka, Understanding Hydroxyl Radical (˙OH) Generation Processes in Photocatalysis, ACS Energy Lett., 2016, 1(2), 356–359.
36. Q. Zhang, G. Xie, M. Duan, Y. Liu, Y. Cai, M. Xu, K. Zhao, H. Tai, Y. Jiang and Y. Su, Zinc Oxide Nanorods for Light-Activated Gas Sensing and Photocatalytic Applications, ACS Appl. Nano Mater., 2023, 6(19), 17445–17456.
37. M. Bai, R. Chen, X. Liu, H. Li, J. Li, H. Huang, M. Song, Q. Zhang, Y. Su, H. Wang, M. Xu and G. Xie, An effective strategy for synthesizing high-performance photocatalyst by recycling the graphite target wastes, J. Environ. Chem. Eng., 2024, 12(5), 113872.
38. H. D. Tran, D. Q. Nguyen, P. T. Do and U. N. P. Tran, Kinetics of photocatalytic degradation of organic compounds: a mini-review and new approach, RSC Adv., 2023, 13(25), 16915–16925.
39. N. M. Mahmoodi, M. Arami, N. Y. Limaee and N. S. Tabrizi, Kinetics of heterogeneous photocatalytic degradation of reactive dyes in an immobilized TiO 2 photocatalytic reactor, J. Colloid Interface Sci., 2006, 295(1), 159–164.
40. E. Padmini and L. R. Miranda, Nanocatalyst from sol–sol doping of TiO 2 with Vanadium and Cerium and its application for 3, 4 Dichloroaniline degradation using visible light, Chem. Eng. J., 2013, 232, 249–258.
41. H. Yang, T. An, G. Li, W. Song, W. J. Cooper, H. Luo and X. Guo, Photocatalytic degradation kinetics and mechanism of environmental pharmaceuticals in aqueous suspension of TiO 2: a case of β-blockers, J. Hazard. Mater., 2010, 179(1), 834–839.
42. N. M. Flores, U. Pal, R. Galeazzi and A. Sandoval, Effects of morphology, surface area, and defect content on the photocatalytic dye degradation performance of ZnO nanostructures, RSC Adv., 2014, 4(77), 41099–41110.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta00833f

---