Lavender-Derived ZnO/Biochar for Photocatalytic Degradation of Doxycycline and Paracetamol

Abstract

The increasing release of pharmaceutical pollutants, particularly antibiotics and analgesics, into aquatic environments poses a significant environmental challenge and necessitates sustainable removal strategies. In this study, lavender-derived biochar was produced by pyrolysis at 450 and 650 °C and subsequently modified with Zn²⁺ (3 and 5 mmol) via a solvothermal method. The resulting materials were evaluated as photocatalysts for the degradation of doxycycline and paracetamol in distilled water under UV-A irradiation. Structural and optical characterization (SEM–EDS, XRD, PL, FTIR) was conducted to elucidate structure–performance relationships relevant to photocatalytic activity. The sample pyrolyzed at 450 °C and modified with 5 mmol Zn²⁺ exhibited the highest photocatalytic performance, achieving degradation efficiencies of 62.78% for doxycycline (k = 0.0032 min⁻¹) and 75.19% for paracetamol (k = 0.0113 min⁻¹). The results demonstrate that controlled Zn incorporation into lavender-derived biochar enhances photocatalytic performance and highlight the role of synthesis parameters in governing catalytic behavior. This work underscores the potential of agro-waste-derived biochar as a functional matrix in sustainable photocatalytic systems.

1. Introduction

The continuous contamination of aquatic systems with pharmaceutical residues has emerged as a critical environmental concern, posing significant risks to water quality, ecosystem stability, and human health [1]. In recent decades, the production and consumption of medicinal products have substantially increased alongside advancements in healthcare. Antibiotics are among the most widely used compounds, with their consumption growing significantly [2]. Due to incomplete metabolism, many active substances are released into aquatic systems, where they are regarded as emerging pollutants. A substantial fraction is excreted via urine or feces and enters wastewater streams, while additional sources include hospital and industrial effluents, improper disposal of medications, and insufficient removal during conventional treatment processes [1]. Even at low concentrations, these residues can adversely affect aquatic organisms, disrupt microbial communities, and contribute to the development of antibiotic resistance [3,4,5].

Currently, antibiotics and analgesics are frequently detected in aquatic environments; thus, they are considered emerging contaminants of emerging concern [1,2,3,4,5,6,7,8]. Tetracycline-type antibiotics have been reported in surface waters and wastewater at concentrations typically ranging from <0.01 to 10 µg/L, depending on the region and proximity to emission sources [9]. Similarly, paracetamol is one of the most widely detected analgesics, typically found at concentrations from tens of ng/L to several µg/L in wastewater effluents. Owing to differences in molecular structure and stability, these compounds exhibit distinct persistence and degradation behaviors in aquatic systems. Doxycycline is known to strongly interact with dissolved ions and suspended particles, forming complexes that hinder its removal [10]. Therefore, doxycycline and paracetamol were selected as representative pharmaceuticals with different structural complexity and environmental behavior, making them suitable model compounds for evaluating the ability of new materials to remove chemically diverse contaminants from water [1]. Because conventional water-treatment methods are often ineffective in eliminating pharmaceutical pollutants, increasing attention has been directed toward more sustainable and cost-effective approaches, such as biochar-based adsorbents and photocatalytic processes [11,12].

Biochar is a carbon-rich material produced through the thermochemical conversion of biomass and has attracted growing interest in wastewater treatment applications. When derived from agricultural residues, biochar represents a low-cost and environmentally sustainable option that can combine adsorption and catalytic properties, aligning with circular economy principles [12,13]. Lavender biomass constitutes an abundant agricultural residue generated during cultivation and processing. Its lignocellulosic composition makes it suitable for pyrolytic conversion into carbon-rich materials. However, un-modified biochar typically exhibits limited photocatalytic activity [14,15,16], necessitating further modification. The performance of biochar is closely linked to its structural characteristics, including aromatic carbon framework development, surface functional groups, and porosity, which collectively influence its reactivity and environmental performance. These properties are strongly affected by pyrolysis temperature [17,18,19]. Higher temperatures generally promote increased aromaticity and structural stability, which may alter adsorption behavior and electron-transfer pathways relevant to photo-catalytic processes [20,21].

To further enhance photocatalytic performance, biochar can be modified by incorporating metal ions. Among the various approaches, Zn²⁺ modification has been widely investigated due to its cost-effectiveness and the formation of ZnO phases with photo-catalytic activity [22,23,24,25]. ZnO is an inexpensive, stable, and non-toxic semiconductor well established as an effective photocatalyst [26,27]. However, excessive zinc loading may lead to particle agglomeration and partial blockage of active sites, reducing accessibility and limiting photocatalytic efficiency [28]. Numerous studies have reported ZnO-supported biochars derived from different agricultural residues, such as banana peels, pistachio shells, palm kernel shells, and other lignocellulosic wastes, demonstrating enhanced removal of organic pollutants through combined adsorption and photocatalysis [29,30,31]. In many cases, the primary emphasis has been placed on overall degradation efficiency rather than systematic evaluation of the influence of pyrolysis temperature on ZnO crystallinity, electronic properties, and degradation kinetics [32,33,34]. Furthermore, comparative evaluation of photocatalytic behavior toward structurally distinct pharmaceutical contaminants under controlled conditions remains limited. Establishing clear correlations between synthesis parameters, structural characteristics, electronic behavior, and catalytic performance is therefore essential for the rational design of sustainable photocatalysts beyond empirical optimization of biomass precursors.

In this context, the present study aims to investigate the photocatalytic performance of Zn-modified lavender-derived biochar, focusing on how pyrolysis temperature and controlled Zn loading influence catalytic behavior toward pharmaceutical degradation. Structural and surface analyses are employed to elucidate structure–performance relationships rather than to comprehensively characterize the biochar itself.

2. Materials and Methods

2.1. Materials

Doxycycline (DC, C₂₂H₂₄N₂O₈, λ_max = 275 and 370 nm) and Paracetamol (PCA, C₈H₉NO₂, λ_max = 243 nm) and from Teva (Sofia, Bulgaria) were selected as model pollutants because of their widespread use and possible environmental effects. Distilled water was utilized in the photocatalytic experiments to demonstrate how an analgesic breaks down in the presence of various contaminants, just as it does in the environment.

All chemicals used for the modification of the biochar, including nitric acid (HNO₃, ≥65%, analytical grade), zinc acetate dihydrate (Zn(CH₃COO)₂·2H₂O, ≥99%, analytical grade), and absolute ethanol (C₂H₅OH, ≥99.8%), were purchased from commercial suppliers and used without further purification (Fluka, Buchs, Switzerland). Zinc acetate dihydrate was used as the Zn precursor and dissolved in ethanol under magnetic stirring to obtain the desired concentration. The resulting solution was then added to the biochar and subjected to solvothermal treatment at 120 °C for 12 h in a Teflon-lined stainless-steel autoclave.

2.2. Preparation of Biochar Samples

The lavender-derived biochar samples used in this study were prepared following previously established procedures [35,36,37]. Briefly, air-dried lavender biomass was carbonized at 450 °C and 650 °C under a nitrogen (N₂) atmosphere (N₂ flow rate 60 mL/min). The obtained biochar was subsequently modified with Zn²⁺ ions via a solvothermal method. Zinc acetate dihydrate (Zn(CH₃COO)₂·2H₂O) was used as the Zn precursor. The amount of precursor was calculated to introduce either 3 or 5 mmol of Zn²⁺ into a total suspension volume of 30 mL. The biochar was dispersed in an ethanolic solution containing zinc acetate and nitric acid, stirred at room temperature for 30 min, and transferred into a Teflon-lined autoclave (Shanghai, China). The solvothermal treatment was carried out at 120 °C for 12 h, according to [36]. After cooling, the samples were washed with distilled water and ethanol, dried at 40 °C for 6 h, and finally used for photocatalytic experiments.

2.3. Characterization Methods

Proximate analysis (moisture, volatile matter, and ash; wt.%) was performed according to ISO standard methods [37], summarized in

Table 1. Physicochemical properties of lavender biomass (adapted from [36]).

Ultimate Analysis, wt. %, d.b.		Ash Analysis (wt. %, d.b.)		Proximate Analysis, wt. %
C	41.14	CaO	14.67	Moisture, r.	7.15
H	6.17	SiO2	8.56	Volatiles, d.b.	68.34
N	1.80	K2O	9.50	Ash, d.b.	13.36
S	2.01	MgO	3.72	Fixed carbon, d.b. 1	11.15
O 1	28.27	Al2O3	2.60
HHV, d.b., MJ/kg	17.45	Fe2O3	1.23

¹ Calculated by difference; Abbreviations: d.b.—on a dry basis; r.—as received.

. Fixed carbon (FC) was calculated by difference as FC = 1 (Wt + A + V), where W_t, A and V denote moisture, ash, and volatile matter contents. Ultimate analysis (C, H, N, S) was carried out using an Elemental Analyzer EuroVector EA 3000 (EuroVector S.p.A., Milan, Italy). The elemental composition of the ash was determined by Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES Telledyne Leeman Labs using Mason, OH, USA). The surface morphology of pristine and Zn²⁺ modified biochar samples was examined by scanning electron microscopy (SEM, Hitachi TM4000, Krefeld, Germany) operated at an accelerating voltage of 10 kV. Mean particle size values in Table 2 were obtained from SEM images using ImageJ software (version 1.54g; National Institutes of Health, Bethesda, MD, USA). Feret diameters were measured for at least 100 manually outlined particles per sample, and the mean values with standard deviations were calculated. Elemental composition and spatial distribution were analyzed using energy-dispersive X-ray spectroscopy (EDS, Quantax 200, Bruker, Berlin, Germany). The X-ray diffraction (XRD) patterns were recorded using a Siemens D500 diffractometer (Karlsruhe, Germany) with CuKα radiation (λ = 1.5406 Å) in the 2θ range of 30–70°, to determine the crystalline structure and phase composition of the samples. The average crystallite size was estimated using the Scherrer equation. Rietveld refinement of the XRD data was performed using PowderCell 2.4 software [28]. Photoluminescence (PL) spectra were used to assess the powder of charge carrier recombination, using fluorescence UV-vis spectrophotometer (Cary, Eclipse, Agilent Technologies Inc., Santa Clara, CA, USA; Mississauga, ON, Canada). The functional groups (e.g., OH, C=O, C–O, C=C, Zn–O) were identified using FTIR spectroscopy. The spectra were measured using a Cary 630 spectrometer (Agilent Technologies Inc., Santa Clara, CA, USA) equipped with a Diamond-ATR (Attenuated Total Reflectance) accessory.

2.4. Photocatalytic Experiments

Photocatalytic experiments were carried out in a cylindrical glass reactor (200 mL) equipped with a magnetic stirrer operating at 500 rpm. The catalyst concentration was 1 g/L in 50 mL of solution. The UV-A lamp irradiance at the solution surface was 1.2 mW/cm², measured using a calibrated radiometer. The distance between the light source and the reactor was 5 cm. A UV-A lamp (36 W, λ = 315–400 nm) was used as the irradiation source. All experiments were performed at room temperature (23 ± 2 °C).

Prior to irradiation, the suspension containing the photocatalyst and the drug solution was stirred at 500 rpm in the dark for 30 min to establish an adsorption–desorption equilibrium. During irradiation, aliquots (2 mL) were withdrawn at regular time intervals, filtered, and analyzed using a UV–Vis spectrophotometer (Evolution 300, Thermo Scientific, Waltham, MA, USA). The initial pH of the solution was 6.5 (adjusted/not adjusted), and the final pH after the photocatalytic test was 5.5. The degradation of doxycycline and paracetamol was monitored at 275 nm and 243 nm, respectively, in order to determine the relative amount of antibiotic in the treated solution.

To verify the establishment of the adsorption–desorption equilibrium, the concentration of doxycycline and paracetamol was monitored during the 30 min dark stirring period prior to irradiation. Aliquots were withdrawn at regular intervals (5, 10, 15, 20, and 30 min), filtered, and analyzed using UV–Vis spectroscopy. In addition, photolysis control experiments were performed under identical UV-A irradiation conditions in the absence of photocatalyst in order to evaluate the contribution of direct photodegradation.

To identify the main reactive species involved in the photocatalytic degradation process, scavenger experiments were performed. Isopropyl alcohol (IPA) and ascorbic acid (AA) were used as scavengers for hydroxyl radicals (OH^•) and superoxide radicals (^•O₂⁻), respectively. Each scavenger (6 mg) was added separately to 50 mL of the reaction mixture before UV irradiation, and the photocatalytic degradation behavior was subsequently evaluated under identical experimental conditions. The total organic carbon (TOC) was investigated for the treated drug solutions, using the Shimadzu TOC-L system (Shimadzu Corporation, Kyoto, Japan).

3. Results and Discussion

3.1. Physicochemical Properties of Lavender

Physicochemical properties of the air-dried biomass are summarized in Table 1. Herein, only the major ash-forming oxides are reported for brevity; detailed composition is available in [36]. The results confirm the biomass’s carbon-rich character and typical mineral composition for lignocellulosic biomass, providing context for subsequent photocatalytic analysis [17,38].

3.2. Structure and Morphology of Pure and Zn²⁺ Modified Biochar Lavender

The structural morphology of each synthetic sample was examined using SEM. Morphological characteristics play a major role in the photocatalytic degradation of organic pollutants in solution. Among the samples, the L/Zn²⁺ (5 mmol, 450 °C) composite was selected based on the maximum photocatalytic degradation, in comparison with other manufactured composites such as L/Zn²⁺ (3 mmol, 450 and 650 °C), carbonized lavender, and air-dried lavender, which are covered later in the section on photocatalysis.

Figure 1 shows SEM images of air-dried solid biomass, and biochar carbonized lavender at two different temperatures: 450 and 650 °C (Figure 1a–c). Figure 1a shows the surface morphology of a pure lavender (L, non-carbonized) sample, with clear aggregates of different forms. As shown in Figure 1b,c, the carbonized sample has a granular shape with smooth surfaces that serve as loading sites for ZnO nanoparticle inclusion. As the pyrolysis temperature increases, larger crystals form and fewer small particles are observed. This undoubtedly affects the catalytic properties and the values of the rate constants.

The SEM structure of Zn²⁺ modified biochar derivatives prepared from a solution containing 3 and 5 millimoles of Zn²⁺ (L/Zn²⁺) is shown in Figure 1d–g. It clearly shows spherical, white ZnO particles over the biochar substrate, indicating the successful formation of a ZnO composite photocatalyst with lavender. The white particles indicate that Zn²⁺ was successfully doped onto the charcoal structure, which is why L/Zn²⁺ was created [29]. The SEM image of L/Zn²⁺ carbonized at 450 °C reveals a rough and irregular surface morphology with visible interparticle voids, which may provide accessible surface sites for adsorption and contribute to the photocatalytic performance [38].

Particle size was determined from SEM images using ImageJ software (National Institutes of Health, Bethesda, MD, USA). A representative region of interest (ROI) was selected from each SEM micrograph, avoiding edge effects and areas with significant particle agglomeration. Particle boundaries were manually outlined, and the Feret diameter was measured for at least 100 particles per sample to ensure statistical reliability. Since the particles exhibit irregular morphology, the Feret diameter was used as an appropriate descriptor of particle size. The mean particle size and standard deviation (SD) were then calculated, and the results are summarized in Table 2.

Table 2. SEM-based particle size of lavender-derived samples as a function of pyrolysis temperature and Zn²⁺ modification.

Biomass Sample	Particle Size ± SD (µm)
Air-dried lavender biomass	45.67 ± 2.15
Lavender carbonized at 450 °C	50.13 ± 1.43
Lavender carbonized at 650 °C	29.60 ± 1.19
L/Zn2+ (3 mmol, 450 °C)	59.28 ± 2.06
L/Zn2+ (3 mmol, 650 °C)	102.25 ± 2.34
L/Zn2+ (5mmol, 450 °C)	31.11 ± 0.91
L/Zn2+ (5 mmol, 650 °C)	31.75 ± 1.67

Table 2. SEM-based particle size of lavender-derived samples as a function of pyrolysis temperature and Zn²⁺ modification.

Biomass Sample	Particle Size ± SD (µm)
Air-dried lavender biomass	45.67 ± 2.15
Lavender carbonized at 450 °C	50.13 ± 1.43
Lavender carbonized at 650 °C	29.60 ± 1.19
L/Zn2+ (3 mmol, 450 °C)	59.28 ± 2.06
L/Zn2+ (3 mmol, 650 °C)	102.25 ± 2.34
L/Zn2+ (5mmol, 450 °C)	31.11 ± 0.91
L/Zn2+ (5 mmol, 650 °C)	31.75 ± 1.67

Representative elemental mappings (C, O, and Zn) of selected Zn-modified biochar samples are shown in Figure 2, whereas Figure 3 displays a similar elemental composition as determined by EDS. The mapping shows that each element is evenly distributed over the sample’s surface. Eleven elements’ compositions were found by the EDS. The results correlate with those reported by other authors [39].

Figure 3 presents the EDS analysis of lavender residue and its carbonized derivatives (biochar) and Zn²⁺-modified biochar. The biochar materials are mainly composed of carbon and oxygen, as observed in Figure 3a–c. The thermal decomposition of the lavender residue during pyrolysis is responsible for the high carbon and oxygen content of the biochar structure. After Zn²⁺ modification (Figure 3d–g), the weight percentages of carbon and oxygen decrease, while the zinc content increases, confirming the successful impregnation of zinc into the biochar matrix.

The quantitative elemental composition obtained from EDS analysis is summarized in

Table 3. The quantitative elemental composition obtained from EDS analysis.

Sample	C (wt%)	O (wt%)	Zn (wt %)	Si (wt%)	Ca (wt%)
Lavender carbonized at 450 °C	63.18	25.58	-	1.48	3.00
Lavender carbonized at 650 °C	54.94	28.37	-	2.42	3.17
L/Zn2+ (3 mmol, 450 °C)	73.56	14.81	5.82	1.68	1.63
L/Zn2+ (5 mmol, 450 °C)	75.58	13.85	7.25	2.33	0.06
L/Zn2+ (3 mmol, 650 °C)	74.14	14.04	9.26	1.87	0.06
L/Zn2+ (5 mmol, 650 °C)	45.43	24.76	22.7	1.42	3.28

to facilitate comparison among samples. The data support the successful in-corporation of Zn in the modified biochars and show an increase in Zn content with in-creasing precursor concentration. Thus, the Zn content was 5.82 wt% and 7.25 wt% for the L450 samples modified with 3 and 5 mmol Zn²⁺, respectively. For the L650 samples, the Zn content increased to 9.26 wt% and 22.7 wt% for the corresponding 3 and 5 mmol modifications, indicating a substantial increase in surface Zn concentration.

The XRD patterns of all samples are shown in Figure 4. When the pyrolysis temperature is increased to 650 °C, the carbon structure of the biochar exhibits a slightly higher degree of short-range ordering, reflected by a somewhat more defined broad diffraction feature in the 20–30° region, characteristic of amorphous carbon structures (JCPDS No. 41-1487). Despite this minor increase in ordering, the material remains predominantly amorphous. Diffraction peaks attributable to crystalline α-SiO₂ (quartz) (011) are observed at 2θ ≈ 26.6°, consistent with JCPDS card No. 46-1045. This feature is present in both samples, and no significant increase in peak intensity with increasing pyrolysis temperature is observed, indicating only minor contributions from crystalline mineral phases [40,41].

After zinc modification, diffraction peaks corresponding to hexagonal wurtzite ZnO are observed for both biochar samples prepared at 450 °C and 650 °C. The main reflections at 2θ ≈ 31.8°, 34.4°, and 36.3° can be indexed to the (100), (002), and (101) planes of ZnO (JCPDS No. 36-1451), confirming the formation of crystalline ZnO on the biochar surface (Figure 4b). For the biochar produced at 450 °C, the ZnO diffraction peaks are broader and less intense, suggesting the formation of smaller and more defective ZnO crystallites, as well as stronger interactions between ZnO and the more disordered carbon matrix. In contrast, the ZnO peaks observed for the biochar prepared at 650 °C are sharper and more intense, indicating improved ZnO crystallinity. This suggests that the more structurally stable biochar matrix formed at higher pyrolysis temperatures promotes the growth of better-ordered ZnO crystallites. The diffraction peak at approximately 26.6°, corresponding to α-SiO₂, is present in both samples and shows little variation, indicating that silicon dioxide originates from the mineral components of the biomass and does not actively participate in the zinc modification process.

To further support the qualitative XRD observations, Rietveld refinement was performed for selected samples, and the main structural parameters are summarized in

Table 4. Rietveld refinement results for Zn-modified lavender-derived biochar samples.

Sample	ZnO (vol.%)	ZnO Crystallite Size (nm)	ZnO Microstrain	Rwp
L/Zn2+ (5 mmol, 450 °C)	43.6	35.2	0.0023	11.7
L/Zn2+ (5 mmol, 650 °C)	54.2	33.1	0.0025	12.9

. The refinement confirms the presence of ZnO and SiO₂ phases in the Zn-modified biochar. The reported phase fractions represent relative percentages within the crystalline components identified by refinement, with values expressed relative to the SiO₂ phase. A higher ZnO phase fraction was observed for the sample pyrolyzed at 650 °C compared to that prepared at 450 °C. The ZnO crystallite size is in the range of approximately 30–35 nm, with only minor differences between the samples, while variations in microstrain suggest differences in defect density and possible interfacial interactions between ZnO and the biochar matrix. The refinement quality indicators indicate a satisfactory fit, considering the partially amorphous nature of the biochar-based materials.

Overall, the XRD and Rietveld refinement results provide a structural explanation for the observed photocatalytic behavior, indicating that the interplay between biochar ordering and ZnO crystallinity, controlled by the pyrolysis temperature, plays a key role in charge transfer processes and, consequently, the degradation efficiency of the studied pharmaceutical contaminants.

Figure 5 presents the photoluminescence (PL) spectra of air-dried lavender biomass, lavender carbonized at 450 °C and 650 °C, and the corresponding Zn²⁺-modified samples (5 mmol). The spectra were recorded at an excitation wavelength of 325 nm, selected to enable excitation across the ZnO band gap and to resolve ZnO-related emission features. PL spectroscopy was employed to evaluate charge-carrier recombination behavior, as emission intensity is commonly associated with the radiative recombination rate of photogenerated electron–hole (e⁻/h⁺) pairs [42,43,44]. For air-dried lavender biomass, the observed emission is attributed to intrinsic organic fluorophores present in lignocellulosic components, rather than semiconductor-related electronic transitions.

Air-dried lavender biomass exhibits the highest emission intensity across the measured spectral range, reflecting the presence of abundant intrinsic organic chromophores. After carbonization at 450 °C and 650 °C, a noticeable reduction in PL intensity is observed. This decrease may be attributed to structural reorganization and increased carbonization degree, which modify the electronic structure of the material and reduce the density of emissive organic groups.

Following Zn incorporation, a further suppression of PL emission is detected. The L/Zn²⁺ (5 mmol, 450 °C) sample exhibits the lowest overall emission intensity among all investigated materials. The reduced PL intensity suggests decreased radiative recombination and potentially improved charge separation in the Zn-modified samples relative to air-dried lavender biomass as well as to the carbonized biomass obtained at 450 °C and 650 °C. The Zn-containing materials exhibit characteristic emission features centered at approximately 387 nm, attributed to excitonic recombination across the ZnO band gap, consistent with the optical properties of hexagonal wurtzite ZnO [26,27]. In addition, a broad visible emission band spanning the 450–550 nm region is observed. This emission is commonly associated with defect-related states in ZnO, which may include oxygen vacancies, zinc interstitials, and other deep-level states within the band gap [25,26,27]. However, PL spectroscopy alone does not allow unambiguous identification of specific defect types. Such defect-related states are frequently reported to contribute to visible luminescence and may influence photocatalytic redox processes [25,26].

In L/Zn²⁺ (5 mmol, 450 °C) and L/Zn²⁺ (5 mmol, 650 °C), defect-related emission is reduced compared to biomass carbonized at 450 °C and 650 °C, respectively. This attenuation suggests that interaction between ZnO species and the carbon matrix formed during carbonization may reduce radiative recombination via defect states or promote interfacial charge transfer. Similar behavior has been reported for Zn-modified biochar, where the conductive phase can act as an electron-accepting component, enhancing charge separation and lowering PL intensity [28,29].

The observed PL quenching can therefore be reasonably attributed to electron transfer from the ZnO conduction band to the conductive carbon matrix. The carbon phase may function as an electron-accepting pathway, promoting spatial separation of charge carriers and suppressing e⁻/h⁺ recombination [42,43,44]. PL analysis provides indirect information on charge-carrier dynamics; however, it does not directly quantify reactive oxygen species (ROS) generation.

Importantly, the PL results are consistent with the radical scavenger experiments performed in this study, which indicated a significant contribution of superoxide radicals (•O₂⁻) in the degradation of both doxycycline and paracetamol. The reduced radiative recombination inferred from PL analysis is consistent with the proposed mechanism involving electron-mediated oxygen reduction reactions. Therefore, the improved photocatalytic performance of L/Zn²⁺ (5 mmol, 450 °C) can be reasonably associated with improved charge separation and facilitated electron transfer between ZnO species and the carbonized lavender matrix.

The PL analysis provides supportive evidence that controlled Zn incorporation combined with optimized carbonization temperature improves interfacial charge-transfer processes, correlating with the observed enhancement in photocatalytic efficiency.

FTIR spectroscopy was used to assess the surface functional groups of lavender-derived biochar before and after Zn²⁺ modification (Figure 6). Broad bands at 3200–3600 cm⁻¹ are attributed to O–H stretching vibrations, while signals in the 1700–1600 cm⁻¹ and 1000–1300 cm⁻¹ regions correspond to C=O, aromatic C=C, and C–O groups typical of lignocellulosic-derived carbon materials [45,46]. Similar spectral features were observed for all samples, with decreasing intensity of oxygen-containing groups at higher pyrolysis temperature, indicating progressive carbonization.

For the Zn²⁺-modified materials, slight attenuation and broadening in the 600–700 cm⁻¹ region were detected, tentatively associated with Zn–O-related vibrations [31,33]. Due to band overlap, specific bonding configurations cannot be conclusively identified by FTIR alone. Residual oxygen-containing functionalities may facilitate interactions between the catalyst surface and pharmaceutical molecules during photocatalytic treatment. The spectra are derived from our previously reported dataset [36] and are discussed here in relation to photocatalytic performance.

3.3. Photocatalytic Degradation of Doxycycline Using Pure and Zn²⁺ Modified Biochar Lavender

The photocatalytic degradation of Doxycycline under UV light irradiation was used to assess the photocatalytic activity of each lavender sample (air-dried biomass and carbonized lavender powders (450 and 650 °C)). The drug concentration was 25 mg/L in every test, and the photocatalytic process was aided by magnetic stirring with a 2.5 cm rod. Doxycycline degradation was tracked using UV–vis spectroscopy by measuring the absorption maxima at 275 and 370 nm. Using manufactured samples under dark and UV light conditions, respectively, the drug’s adsorption and photocatalytic degradation were examined. Prior to the photocatalytic degradation, the dark adsorption–desorption studies were carried out. It anticipates that the photocatalysts, which have adsorption sites for the Doxycycline molecules, will become saturated with adsorbate (drug) and reach an equilibrium state after 15 min of darkness. The air-dried biochar lavender photocatalyst does not exhibit a change in drug concentration following light irradiation (Figure 7a), as it appears as a nearly linear curve over the course of 120 min [47]. Given that there is no discernible photocatalytic breakdown of the medication, this small shift suggests that biochar is a great adsorbent for the removal of doxycycline.

For doxycycline (initial concentration 25 mg/L), a rapid decrease in concentration was observed within the first 15 min, followed by stabilization of the concentration between 15 and 30 min. The variation in concentration after 15 min was less than 3%, indicating that the adsorption–desorption equilibrium was effectively reached before the onset of UV-A irradiation. Extending the dark period beyond 30 min did not result in measurable additional adsorption under the selected experimental conditions. Therefore, 30 min was considered sufficient to ensure equilibrium and was applied consistently in all photocatalytic experiments.

The photodegradation results in Figure 7 reveal that after 120 min, carbonized lavender powder annealed at 450 °C exhibits the fastest degradation rate. The surface morphology of carbonized lavender is likely responsible for its increased efficiency. Smaller particles cover the entire surface, improving carrier participation in redox processes. The carbonized samples have a markedly altered morphology, with particle aggregation evident. Agglomeration may reduce the accessibility of surface-active sites and promote charge carrier recombination, thereby decreasing photocatalytic efficiency. Using the pseudo-first-order model −Ln(C/C₀) = kt, the reaction rate constants (k), depicted in Figure 7b, were calculated. These findings were further supported by the fact that carbonized lavender at 450 °C showed the highest rate constant (k = 0.0032 min⁻¹) at 500 rpm. Using air-dried and carbonized lavender at 450 and 650 °C, the degradation efficiencies D, D (%) = ((C₀ − C)/C₀) × 100, where C and C₀ are at time t and the initial concentration) for the three samples were 0%, 32.75%, and 21.59%, respectively. As the temperature rises to 450 °C, the photocatalytic efficiency increases as a result of the development of smaller particles, or active centers, which improve the interaction between the drug molecules and the catalyst and accelerate the photocatalytic activity. Although higher temperatures provide more energy to molecules (increasing the rate of reactions), they can shift the equilibrium in a direction that is not favorable for the formation of products or enhance unwanted side reactions that deactivate the catalyst. It has been shown using SEM that at higher temperatures, some of the small particles (which have a high surface area) can coalesce (agglomerate) into larger crystals. This reduces the total active surface area, even though the crystals themselves are larger. Although larger crystals are not always bad, in catalysis, the key factor is the total active surface area and the quality of these centers. Agglomeration and deactivation at high temperatures are often the dominant factors leading to a decrease in activity, even if the individual crystals are larger.

According to reports, biochar offers a variety of surface groups for the enhancement of photocatalytic activity [48]. When compared to pure biochar lavender, the hybrid photocatalysts’ photocatalytic degradation efficiency of doxycycline increases dramatically as Zn²⁺ concentrations rise. The increased degradation of doxycycline is caused by the production of additional electron–hole-based radicals (superoxide and hydroxyl) in L/Zn²⁺, which interact with drug molecules on the surface of L/Zn²⁺. Figure 8 shows that of all the developed photocatalysts, L/Zn²⁺, 5 mmol produced the maximum photocatalytic degradation rate of doxycycline, followed by L/Zn²⁺, 3 mmol, and biochar lavender. Doxycycline can be photocatalytically degraded at a rate of up to 62.78% using L/Zn²⁺, 5 mmol (the highest rate of any composite, i.e., L/Zn²⁺, 3 mmol (53.85%), carbonized at 450 °C). The samples that were carbonized at 650 °C showed the same pattern. Drug degradation is, nevertheless, less common. This is because of the reduction in active centers and the aggregation of crystallites. Thus, it can be said that during the fabrication of L/Zn²⁺ samples, the photocatalytic degradation of doxycycline increased as the loading amount of ZnO on the surface of biochar lavender increased. The trend of increased photocatalytic efficiency at greater Zn²⁺ concentrations is confirmed by Figure 5b, which shows the matching rate constant values for both photocatalysts.

To evaluate the possible contribution of direct photodegradation, control experiments were performed under identical UV-A irradiation conditions (36 W, 315–400 nm, 23 ± 2 °C) in the absence of photocatalyst. The results showed that doxycycline exhibited negligible degradation (<1%) after 120 min of irradiation. This behavior is consistent with the limited absorption of doxycycline in the UV-A region compared to shorter UV wavelengths. Therefore, the concentration decrease observed in the presence of Zn-modified lavender biochar can be attributed predominantly to photocatalytic activity rather than direct photolysis.

The carbonaceous matrix does not act merely as an inert support but contributes multifunctionally to the photocatalytic system. Its porous structure enhances the adsorption of pharmaceutical molecules, promoting their proximity to ZnO active sites and facilitating adsorption-assisted photocatalysis. In addition, the biochar framework improves ZnO dispersion and may reduce particle aggregation. The conductive carbon matrix can also favor interfacial charge transfer and contribute to reduced electron–hole re-combination, as suggested by the comparative PL results. Furthermore, the use of agro-waste-derived biochar provides a sustainable and low-cost support consistent with circular economy principles. Rather than solely maximizing the intrinsic activity of ZnO, the composite approach aims to integrate adsorption capacity, photocatalytic functionality, and material sustainability into a single system. This multifunctional behavior should be considered when interpreting the overall photocatalytic performance of the developed materials. Based on these structural and interfacial characteristics, the possible photocatalytic mechanism is discussed below.

3.4. Photocatalytic Mechanism

Figure 9 illustrates the proposed interfacial charge-transfer and surface-mediated photocatalytic mechanism of the ZnO–biochar composite under UV-A irradiation. The scheme integrates the electronic band structure of ZnO with the heterogeneous surface reactions occurring at the ZnO–biochar interface [49].

Upon irradiation with photons of energy equal to or higher than the band gap of ZnO, electrons are excited from the valence band (VB) to the conduction band (CB), generating electron–hole pairs:

ZnO + hν → e⁻(CB) + h⁺(VB)

(1)

The positions of the conduction and valence bands allow thermodynamically favorable redox reactions. The CB edge of ZnO is sufficiently negative to reduce molecular oxygen to superoxide radicals, while the VB edge is sufficiently positive to oxidize water or surface hydroxyl groups to hydroxyl radicals. Thus, ZnO possesses the intrinsic redox potential required for reactive oxygen species (ROS) generation.

A key feature of the composite system is the presence of the ZnO–biochar interfacial region. As depicted in Figure 8, part of the photogenerated electrons migrates from the ZnO conduction band toward the biochar phase. This interfacial electron transfer suppresses electron–hole recombination, as schematically indicated by the inhibited recombination pathway. The conductive carbon matrix acts as an electron mediator or sink, promoting spatial separation of charge carriers and prolonging their lifetime. This enhanced charge separation increases the probability of surface redox reactions.

In parallel, biochar contributes through its abundant oxygen-containing surface functional groups (–OH, –COOH, C=O), which serve as adsorption sites for the target pollutant molecules. An adsorption equilibrium is established between dissolved species and surface-bound molecules:

Drug (aq) ⇌ Drug (ads)

(2)

This adsorption step is mechanistically important, as the degradation is expected to predominantly occur on surface-adsorbed molecules. The preconcentration of the pollutant at the ZnO–biochar interface enhances the efficiency of the catalytic process.

At the catalyst surface, the photogenerated electrons reduce adsorbed oxygen molecules:

O₂(ads) + e⁻ → •O₂⁻

(3)

Simultaneously, valence-band holes oxidize adsorbed water or hydroxide species:

H₂O(ads) + h⁺ → •OH + H⁺

(4)

OH⁻(ads) + h⁺ → •OH

(5)

The generated superoxide (•O₂⁻) and hydroxyl (•OH) radicals react predominantly with the adsorbed pollutant molecules:

•O₂⁻/•OH + Drug (ads) → intermediates → mineralization products

(6)

Importantly, these reactions occur at well-defined surface active sites, as indicated in the schematic representation. Therefore, the overall degradation process is governed by surface-controlled heterogeneous redox reactions rather than homogeneous radical processes in the bulk phase.

The synergistic coupling between adsorption and photocatalysis explains the enhanced performance of the composite compared to pristine ZnO. Biochar does not merely act as a passive adsorbent but actively participates in interfacial charge transfer and facilitates the formation of reactive species. The proposed mechanism is consistent with the observed suppression of charge recombination and the dominant role of superoxide radicals indicated by radical scavenging experiments.

Overall, Figure 9 highlights that the improved photocatalytic efficiency arises from adsorption-enhanced heterogeneous photocatalysis driven by interfacial charge separation and surface-mediated redox reactions at the ZnO–biochar interface.

It should be noted that, due to the presence of the carbonaceous matrix, possible interactions between scavenger molecules and the biochar surface cannot be completely excluded. Therefore, the radical trapping experiments provide indicative rather than definitive evidence regarding the dominant reactive species under the investigated conditions and are interpreted qualitatively. While the present results demonstrate effective degradation of the parent compounds under UV-A irradiation, detailed identification of intermediate products and full elucidation of the degradation pathway were beyond the scope of this study.

3.5. Photocatalytic Degradation of Paracetamol Using Pure and Zn²⁺ Modified Biochar Lavender

Ultimately, the biochar samples’ electronic structure was successfully adjusted using Zn²⁺ modification. Furthermore, we were inspired to use this test for another organic pollutant in the presence of UV light due to their exceptional efficacy. Using the same method and catalyst type, the powders’ photocatalytic activity for decolorizing doxycycline and paracetamol was examined. The photocatalytic properties of pure lavender and lavender/Zn²⁺ powders for Paracetamol degradation under UV light irradiation are contrasted in Figure 10. The drug had a starting concentration of 25 ppm. The photocatalytic tests demonstrated pseudo-first-order kinetics for the drug’s decolorization by the powders when exposed to UV light. The rate constant of photocatalysis, k, is represented by the slope of the logarithmic scale linear fits. The negligible degradation observed in the photolysis control further confirms that the enhanced removal efficiency is attributed to the synergistic adsorption–photocatalysis effect of the Zn-modified biochar composites. Similar photolysis control experiments were conducted for paracetamol under identical UV-A irradiation conditions in the absence of photocatalyst. The results indicated negligible degradation (<1%) after 120 min of irradiation. Since paracetamol shows weak absorption in the applied UV-A spectral range, direct light-induced decomposition is minimal. Consequently, the enhanced degradation efficiency observed in the presence of Zn-modified biochar composites confirms that the removal process is mainly governed by photocatalytic mechanisms.

The L/Zn²⁺, 5 mmol, carbonized at 450 °C samples had the highest photocatalytic activity (k = 0.0113 min⁻¹) in the photocatalytic processes conducted in the presence of UV light, as Figure 10 illustrates. The powdered L/Zn²⁺, 3 mmol, carbonized at 450 °C was more efficient (k = 0.0091 min⁻¹) than the pure lavender (k = 0.0051 min⁻¹). As confirmation of the rate constants are the values of the percentage of degradation of organic pollutants. The modified powders had a greater activity (Paracetamol—L/Zn²⁺, 3 mmol = 66.25%, L/Zn²⁺, 5 mmol = 75.19%; Doxycycline—L/Zn²⁺, 3 mmol = 53.85%, L/Zn²⁺, 5 mmol = 62.78%) than the pure samples in the photocatalytic tests pertaining to the breakdown of paracetamol and doxycycline. This finding supported Zn²⁺’s beneficial impact on biochar’s photoinduced efficiency.

Pure lavender powder and Zn²⁺ modified powder showed faster degradation of paracetamol compared to doxycycline due to the formation of stable intermediates by the reaction of OH^• with aromatic structure during the photocatalytic process. Paracetamol has a relatively simple aromatic structure with amide and hydroxyl groups. This makes it an easy target for hydroxyl radicals, which are the main agents in catalytic degradation. In contrast to Doxycycline, which is a tetracycline antibiotic with four fused rings and multiple functional groups. This structural complexity often leads to the formation of stable intermediates that delay complete mineralization [50,51,52,53].

The radical scavenger assay was used to assess the involvement of hydroxyl and superoxide radicals. Figure 11 shows the data. By adding ascorbic acid (AA) and isopropyl alcohol (IPA), scavengers which capture the corresponding reactive species, drug degradation by superoxide and hydroxyl radicals was evaluated [54,55]. The degradation efficiency of doxycycline and paracetamol was not significantly altered by the addition of IPA. AA, however, showed greater inhibition, suggesting that superoxide radical generation plays a more significant role in the degradation rate of both medications. When AA and IPA were added, the degradation rates for doxycycline were 51. 13% and 46. 34%, and for paracetamol 57. 57.72% and 53. 28%, using L/Zn²⁺, 5 mmol 450° C, as shown in Figure 11. These findings also indicate that paracetamol degrades more quickly than doxycycline. Drug degradation depends on superoxide radicals, as illustrated in Figure 8, with the AA scavenger causing the least degradation.

Figure 12 presents the recyclability performance of pristine and Zn-modified lavender-derived photocatalysts over three consecutive degradation cycles under identical experimental conditions. For the optimized sample (L/Zn²⁺, 5 mmol, 450 °C), the degradation efficiency toward both doxycycline and paracetamol decreased by approximately 2% after three cycles. This minor decline indicates a slight reduction in catalytic activity upon reuse, while the overall photocatalytic performance remains largely preserved.

The small decrease in efficiency may be attributed to partial blockage of surface-active sites by adsorbed intermediates, limited surface fouling, or minor aggregation during repeated operation. Nevertheless, no pronounced loss of photocatalytic activity was observed within the tested cycles, suggesting that the ZnO–biochar composite exhibits good short-term operational stability under UV-A irradiation.

These results demonstrate satisfactory reusability of the developed materials under the investigated laboratory conditions. Further long-term stability evaluation under extended irradiation will be addressed in future work.

The photocatalytic activity of the optimized catalyst L/Zn²⁺ (5 mmol, 450 °C) was systematically investigated for the degradation of doxycycline and paracetamol under UV irradiation. All experiments were performed under continuous magnetic stirring to ensure homogeneous suspension of the catalyst particles and to minimize mass-transfer limitations. The initial concentration of each pharmaceutical compound was fixed at 25 mg/L, allowing a direct comparison of degradation efficiency.

The photocatalytic process was monitored using UV–Vis spectroscopy by recording the time-dependent evolution of the characteristic absorption bands of doxycycline and paracetamol. Changes in the intensity and profile of these bands were used to evaluate the degradation progress. A gradual decrease in the main absorption peaks was observed during irradiation, indicating the breakdown of the parent molecules rather than simple adsorption onto the catalyst surface. The absence of significant peak shifts suggests molecular transformation and progressive structural decomposition of the pharmaceuticals.

The spectral evolution provides insight into the catalytic efficiency of the L/Zn²⁺ (5 mmol, 450 °C) material throughout the reaction. The steady decline in absorbance over time reflects the effective generation of reactive oxygen species and the sustained activity of the catalyst under UV exposure. The complete set of UV–Vis spectra recorded during the degradation process over the pure L/Zn²⁺ (5 mmol) sample is presented in Figure 13, illustrating the temporal decrease in the characteristic absorption bands for both target compounds.

The mineralization efficiency of doxycycline and paracetamol was assessed by total organic carbon (TOC) analysis using the optimized catalyst L/Zn²⁺ (5 mmol, 450 °C). The results indicate higher TOC removal for paracetamol (51.27%) compared to doxycycline (40.13%) in distilled water, suggesting a more extensive oxidative conversion of paracetamol under the applied photocatalytic conditions. When compared with degradation efficiencies determined by UV–Vis spectroscopy (Figure 13), the TOC removal values were consistently lower. This discrepancy highlights that the disappearance of the parent compound does not necessarily correspond to complete mineralization. While UV–Vis analysis monitors the decay of the characteristic absorption bands of the original molecules, TOC reflects the total organic carbon content in solution and therefore provides quantitative information on the extent of organic carbon mineralization. The observed difference between UV–Vis degradation and TOC removal indicates the formation of partially oxidized intermediate species during the photocatalytic process. These intermediates contribute to the residual organic carbon content and may exhibit higher structural stability, especially in the case of doxycycline due to its polycyclic and multifunctional structure. From a mechanistic perspective, photocatalytic oxidation proceeds through successive radical-mediated steps involving reactive oxygen species, primarily superoxide (•O₂⁻) and hydroxyl radicals (•OH). Each intermediate compound possesses distinct oxidation potentials and may require multiple electron-transfer steps before complete mineralization is achieved. Therefore, the TOC analysis indicates that mineralization proceeds through sequential oxidative pathways and may be kinetically limited by the stability of intermediate products. This behavior is consistent with the multi-step radical oxidation mechanism proposed for ZnO-based photocatalytic systems [56,57].

For comparison with previously reported ZnO/biochar photocatalysts,

Table 5. Comparative summary of recent ZnO/biochar-based photocatalysts for pharmaceutical degradation.

Type of Catalyst	Light Source	Target Contaminant	Catalyst Dosage (g·L−1)	Removal Efficiency (%)	Rate Constant k (min−1)	Study (Ref.)
Biochar derived from lavender residue (3 & 5 mmol Zn2+; 5.8–22.7 wt% Zn, EDS)	UV-A	Doxycycline	1.0	62.78	0.0032	This work (lavender)
		Paracetamol	1.0	75.19	0.0113
Fixed Zn precursor; varying biochar content (1–5 wt%)	UV-A (365 nm)	Carbamazepine, Ibuprofen	0.1–1.0	95–99	NR	[32]
ZnO/PiC (pine-derived biochar)	Visible (Xe lamp)	Metronidazole	0.2	97.1	0.083	[33]
ZnO/g-C3N4/biochar	Visible (>420 nm)	Doxycycline	0.5	98.9	0.069	[34]
ZnO/biochar derived from: sugarcane bagasse	UV-A, 125 W high-pressure mercury lamp	Sulfamethoxazole	0.5	99.3%	NR	[58]

summarizes selected studies addressing pharmaceutical degradation. As shown, although several systems exhibit high removal efficiencies under UV or visible irradiation, direct comparison remains challenging due to variations in catalyst composition, catalyst dosage, light source and irradiation intensity, initial pollutant concentration, and reactor configuration. Degradation efficiency and apparent rate constants in heterogeneous photocatalysis are strongly influenced by these operational parameters. Therefore, the comparison presented here aims to provide a general performance context rather than a direct one-to-one benchmarking. While some multi-component systems report higher apparent rate constants, the present study emphasizes a simpler binary ZnO/biochar system with controlled synthesis parameters, enabling systematic evaluation of structure–performance relationships. Moreover, quantitative reporting of Zn content and its correlation with photocatalytic kinetics remain insufficiently addressed in many previous studies.

4. Conclusions

This work demonstrates that the photocatalytic performance of Zn-modified lavender-derived biochar is influenced by both pyrolysis temperature and Zn incorporation. The enhanced activity observed for the material prepared at 450 °C suggests that moderate carbonization provides favorable structural conditions for ZnO interaction within the carbon matrix, whereas higher temperatures may promote aggregation effects that limit catalytic efficiency. The results indicate that catalytic behavior depends not only on Zn loading but also on the structural characteristics of the biochar support, which appear to affect reactive species generation under UV-A irradiation. The present study was conducted under controlled laboratory UV-A conditions. Future investigations under visible-light irradiation and extended operational stability tests, including Zn leaching evaluation, will be important for assessing practical implementation.

Overall, the findings provide insight into how controlled synthesis parameters can be used to tune agro-waste-derived biochar systems, supporting the development of more sustainable photocatalytic materials for water treatment.