Enhanced Photocatalytic Degradation of Ciprofloxacin Under Natural Sunlight Using a Waste-Derived Carbon Dots–TiO₂ Nanocomposite

Abstract

The presence of emerging organic contaminants in water and effluents, including antibiotics, poses significant environmental and health risks. Moreover, while photocatalysis is a promising approach for their removal, the inefficient utilization of natural sunlight by common photocatalysts limits its large-scale use. This work demonstrates the enhanced sunlight-driven photodegradation of the antibiotic Ciprofloxacin (CIP) using a nanocomposite composed of carbon dots (CDs) and TiO₂ (NC_50:50). The CDs were obtained from corn stover, a major agricultural waste product. Initial testing was performed under artificial solar radiation: CIP was virtually fully degraded within 20 min, with a rate constant of 0.2372 min⁻¹ and a 217% enhancement of catalytic activity over commercial TiO₂. Validation under real-world irradiation conditions was subsequently made by performing photocatalytic assays under natural sunlight on different days under diverse meteorological conditions. The performance of NC_50:50 was retained, degrading CIP within 30 min under natural conditions. Notably, while degradation by-products were identified under both artificial and natural sunlight, they were subsequently photodegraded by the nanocomposite under these conditions. This enhanced performance was attributed to a combination of effects resulting from CDs’ incorporation, namely, improved absorption of visible light, enhanced charge separation, and increased specific surface area. Furthermore, the addition of CDs resulted in changes in the reactive species generation profile, which can alter the available degradation pathways. Thus, this study provides insight that can be useful for strategies aimed at the rational design of sunlight-active TiO₂-based photocatalysts with tunable surface reactivity.

1. Introduction

Water is an essential resource for humanity, and thus, its contamination by persistent organic pollutants, such as antibiotics, is a significant concern, especially when we consider that conventional water treatment techniques are unable to efficiently remove this type of pollutant from water supplies [1]. The ineffectiveness stemming from the complex structure of antibiotics allows these pollutants to reach water streams through the effluents of wastewater treatment plants, resulting in their widespread presence in countries worldwide [2]. Batt et al. detected ciprofloxacin (CIP), sulfamethoxazole, trimethoprim, tetracycline, and clindamycin in wastewater effluents from three treatment plants in New York State, USA [3]. Liu et al. found that quinolone antibiotics were major pollutant factors in Chinese lakes [4]. Rodriguez-Mozaz et al. reported the detection of ciprofloxacin in several European countries, with concentrations as high as 1435.5 ng L⁻¹ being detected in Portugal [5].

The widespread presence of antibiotics in aquatic systems is further exacerbated by the absence of policies regulating their consumption in several countries, leading to excessive use. Ultimately, they reach wastewater treatment plants and are subsequently discharged into environmental water, where they range from the ng L⁻¹ to the µg L⁻¹ [6]. Additionally, industries such as pharmaceutics, agriculture, healthcare, and livestock farming can also lead to the infiltration of antibiotics into aquatic systems [7].

This widespread presence of traceable amounts of human and veterinary antibiotics poses a significant threat to public health, as it contributes to the development and dissemination of antibiotic resistance [8,9,10]. In aquatic ecosystems, the continuous release of antibiotics facilitates the selection of antibiotic-resistant bacteria and promotes the spread of antibiotic resistance genes, which are subsequently transferred within microbial communities [2]. As a result, bacterial infections become more difficult to treat, leading to higher healthcare costs and increased mortality. Moreover, the accumulation of antibiotics in aquatic environments also exerts toxic effects on non-target organisms, disrupting ecological balance and biodiversity [11]. Given these critical consequences, the effective degradation of antibiotics and their removal from aquatic systems is of paramount importance to mitigate their resulting environmental and public health impacts.

Among the antibiotics present in aquatic systems, we find CIP, a synthetic fluoroquinolone with antibacterial activity against both Gram-positive and Gram-negative bacteria, such as Escherichia coli and Staphylococcus aureus [12]. Regarding CIP, despite its broad-spectrum efficacy and widespread use in aquatic environments, concerns exist regarding its ecological impact and potential contribution to antibiotic resistance. As with most antibiotics, conventional wastewater treatment methods are often ineffective at fully degrading CIP, leading to its accumulation in surface water, groundwater, and sediments.

Considering the limitations and ineffectiveness of conventional wastewater treatment methods regarding removing antibiotics, it is no surprise that the scientific community has focused its efforts on alternative technologies. Photocatalysis is a potentially clean technology that has been widely studied for various applications, utilizing light irradiation as the energy source behind the intended catalytic reaction [13,14,15,16]. It is frequently a less impactful method than conventional alternatives for removing organic pollutants, as it exhibits high redox performance, produces minimal amounts of sludge, and can utilize a clean energy source, such as solar irradiation, to drive the degradation process [17].

Thus far, several photocatalysts have been investigated for environmental remediation and pollutant removal, primarily focusing on metal oxides and sulfides [16,18,19,20]. Among those studied by the scientific community, TiO₂ is particularly important due to its widespread application, cost-effectiveness, and chemical stability. Furthermore, it is already extensively utilized in photocatalytic technologies for environmental remediation, including the degradation of organic pollutants such as pharmaceutical contaminants (e.g., ibuprofen), the photoreduction and removal of toxic heavy metals, and the generation of hydrogen via water splitting [13,14,16,21]. Despite these advantages, TiO₂ presents several limitations that restrict its use in a wider range of applications. First and foremost, it exhibits a wide bandgap (around 3.2 eV) that prevents efficient utilization under sunlight (of which only around 5% is UV radiation) [22,23]. Additionally, TiO₂ suffers from a rapid recombination of photogenerated holes (h⁺) and electrons (e⁻), [24] and presents a poor affinity for organic pollutants [25].

Given these challenges, recent research efforts have focused on optimizing photocatalytic materials via methods using doping with both metals [26] and non-metals [27] and conjugation with alternative materials such as graphene [28]. Due to the increased structural stability of the crystalline framework and the introduction of new energy levels, which extend the light absorption capabilities, applying these strategies may significantly enhance photocatalytic efficiency under visible-light irradiation [26,27,28,29]. However, many of these techniques rely on incorporating potentially toxic metal elements, which, in addition to being expensive, can raise health and environmental concerns. Additionally, such modifications often lead to increased charge carrier recombination, ultimately reducing photonic efficiency and diminishing the overall catalytic performance [30].

An alternative approach to overcoming these limitations is the conjugation of semiconductors with carbon dots (CDs). These nanometric carbon-based particles exhibit spherical or quasi-spherical morphologies, possess either an amorphous or crystalline core, and a functionalized surface [31]. CDs possess several advantageous properties relevant to photocatalysis, including strong photoluminescence, a broadband absorption range that extends into the visible range, high water solubility, and low toxicity [32,33,34,35]. Due to being low cost, easy to prepare, and presenting versatile physicochemical properties, CDs have been tested and employed in various applications, such as sensing, [36] light-emitting devices, [37] photodynamic therapy, [38] and photocatalysis [15,16,39,40]. In the latter, CDs can function as independent photocatalysts or in conjunction with semiconductors, such as TiO₂, to enhance light absorption and improve photonic efficiency, thereby improving overall photocatalytic performance.

The conjugation of CDs with TiO₂ has the potential to enhance the sunlight-driven photocatalytic performance while avoiding the drawbacks associated with doping or coupling with metal-based materials. Unlike these strategies, adding CDs to TiO₂ does not introduce additional toxicity or compromise the photonic efficiency of TiO₂, making it a more environmentally and economically viable alternative [30,41,42]. CDs enhance photocatalytic activity through several mechanisms. For one, CDs can effectively accept photoexcited e⁻ from the conduction band of TiO₂, thereby improving the separation of photoexcited e⁻/h⁺ pairs [43]. Additionally, the broadband absorption range of CDs allows for an absorption range that extends into the visible region, enabling a more efficient utilization of solar irradiation [15,44,45]. Finally, the functionalized surface of CDs increases the affinity of the nanocomposites’ surface toward organic pollutants, enhancing adsorption onto the photocatalyst and facilitating their interaction with active sites. All these factors contribute to improved sunlight-driven photodegradation, making CD-TiO₂ composites a highly effective and sustainable solution for photocatalytic applications [45].

Another significant advantage of using CDs over other methods is their versatility in synthesis, as they can be produced from a diverse range of carbon precursors, including organic waste. This can help to upcycle these residues into valuable nanomaterials. Moreover, life cycle assessment studies indicate that the most impactful stage of the life cycle of nanomaterials is their synthesis, and that the identity of the carbon precursor employed in the synthesis of CDs significantly influences the resulting potential environmental impacts of the process [46,47,48]. By using organic waste such as corn stover instead of commercially available reagents, we can avoid the impacts associated with the preparation of these chemicals and thus reduce the overall impact of the CDs. Corn stover, a major agricultural waste product consisting of the leaves, stalks, and cobs left after corn harvest, represents approximately 50% of the total corn crop yield; yet, only about 10% is repurposed for other applications [49]. Given its abundance, upcycling corn stover into a nanotechnology-based solution for wastewater treatment presents a promising opportunity for both environmental remediation and resource sustainability.

Herein, we report how the efficient and virtually complete sunlight-driven photodegradation of the emerging contaminant CIP can be achieved with a CDs and TiO₂ nanocomposite (NC_50:50), under both artificial and natural irradiation conditions. The CDs were prepared from corn stover, thereby combining circular-economy principles with solar-driven approaches to develop more environmentally sustainable wastewater treatment processes.

While this nanocomposite was characterized earlier [16], its sunlight-driven photocatalytic activity (under real-world irradiation and artificial conditions) and its performance as a catalyst toward the photodegradation of pharmaceutical contaminants were investigated here for the first time. This type of investigation, conducted under real-world conditions, is essential for future large-scale applications of TiO₂-based photocatalysis for (waste)water treatment.

More specifically, when first using a solar simulator for assays in artificial conditions, NC_50:50 was able to virtually fully degrade CIP within 20 min (k of 0.2372 min⁻¹), while showing a 217% enhancement of photocatalytic activity toward commercial TiO₂. Some potential degradation by-products were identified by HPLC-DAD and subsequently shown to be photodegraded by NC_50:50. The photocatalyst was found to be stable over multiple photocatalytic cycles, and scavenger assays identified the reactive species involved. Finally, assays conducted under real-world irradiation conditions with natural sunlight showed that NC_50:50 consistently fully degrades CIP within 30 min, even when prepared on different days under diverse meteorological conditions.

Furthermore, we observed that the addition of CDs led to a shift in the reactive species generation profile, with CIP being degraded by a combination of different reactive species (superoxide anion, h⁺, e⁻, and singlet oxygen). This approach can then enable a strategy to tune the reactivity of TiO₂-based photocatalysts and broaden their application to a wide range of contaminant classes.

2. Results and Discussion

2.1. Description of the Photocatalyst and Prior Characterization

The TiO₂—CDs nanocomposite (NC_50:50) used in this study was previously developed by our group and comprehensively characterized, including by XRD, SEM, XPS, solid-state (ss) NMR, FTIR, PL, and UV-Vis spectroscopy, as reported in [16]. In the present work, this material was reproduced following the previously reported synthesis and evaluated for the first time as a sunlight-driven photocatalyst for the degradation of pharmaceutical contaminants.

NC_50:50 was prepared from TiO₂ and CDs (prepared using corn stover and citric acid (CA) as co-carbon sources, and ethylenediamine (EDA) as N-dopant in a 1:1:2 ratio, respectively) [16]. Both NC_50:50 and commercially available TiO₂, thereon employed as a benchmark, were tested here as photocatalysts for the sunlight-driven photodegradation of CIP in aqueous solution. It should be noted that while NC_50:50 was previously obtained, its photocatalytic performance under sunlight irradiation, either artificial or natural, had not been investigated prior to this work, nor had its application to pharmaceutical contaminants.

Given that the nanocomposite was first developed in a previous study, the results of its previous comprehensive characterization are summarized here for the benefit of the reader, with the full datasets being found in [16].

Regarding optical properties, the introduction of carbon dots (CDs) led to an extended absorption range up to 700 nm [16], enhancing their visible-light absorption capability without changing the bandgap of NC_50:50 (3.26 eV) [16]. By their turn, photoluminescence studies suggested that adding CDs leads to suppressing the recombination of TiO₂’s photoinduced charge carriers [16], due to lower photoluminescence than that measured for sole TiO₂ [16]. Electrochemical impedance spectroscopy further supported this finding by indicating lower electron transfer resistance under irradiation [15,16]. These results indicate that CDs enhance visible-light absorption and facilitate more efficient charge separation [16], ultimately improving the photocatalytic performance of the nanocomposite.

In terms of structural and chemical composition, NC_50:50 was assessed using various characterization techniques, including SEM, XRD, FTIR, XPS, and ss-NMR spectroscopy [16]. SEM analysis revealed that NC_50:50 exhibited a spherical morphology with an average diameter of approximately 92.9 nm. XRD patterns were consistent with anatase-phase TiO₂, while FTIR and XPS analysis identified functional groups such as carbonyl, hydroxyl, and amines, which may influence the nanocomposite’s interaction with pollutants and confirm the presence of the CDs on the TiO₂. ss-NMR spectroscopy provided further insights into the organic structure of the CDs, revealing highly oxygenated aliphatic chains that interact with TiO₂. These findings suggest that the CDs play a crucial role in modifying the nanocomposite’s surface properties, thereby further enhancing its photocatalytic performance and providing a foundation for our work [16].

2.2. Photocatalytic Optimization and Assays

The photocatalytic efficiency of NC_50:50 under simulated sunlight toward CIP was evaluated via RP-HPLC-DAD. Before optimizing the photocatalytic conditions, preliminary tests were conducted with 20 mg of photocatalyst and 20 ppm of CIP under 60 min of irradiation, with samples collected every 15 min. Additionally, the same assay was performed using commercial TiO₂ (<25 nm particle size, anatase phase, from Sigma-Aldrich), which served as a benchmark.

To evaluate NC_50:50’s performance relative to TiO₂, we quantified the percentage of CIP removed over a fixed irradiation period (%DR, Equation (1)) [50] and determined the corresponding rate constant (Equation (2)). The %DR was calculated using the following equation:

$$\mathrm{\%}\mathrm{D}\mathrm{R}= {\displaystyle \frac{{\mathrm{C}}_0-{\mathrm{C}}_{\mathrm{t}}}{{\mathrm{C}}_0}} \times 100$$

(1)

Here, %DR corresponds to the % of CIP removed, C₀ represents the initial CIP concentration, and C_t is the concentration of CIP after an irradiation period of t minutes.

Moreover, the kinetic equation of the degradation follows a pseudo-first-order reaction that can be expressed as follows [51]:

$$-\ln \left({\displaystyle \frac{{\mathrm{C}}_{\mathrm{t}}}{{\mathrm{C}}_0}}\right)=kt$$

(2)

In this case, k corresponds to the estimated reaction rate constant.

The analysis of the initial results obtained revealed that, even prior to optimization, NC_50:50 presented a significantly better CIP photodegradation potential than TiO₂, both in terms of reaction rate constant (0.0536 min⁻¹ vs. 0.0283 min⁻¹, enhancement of 89.4%, depicted in Figure S1a) and degradation extent, with NC_50:50 managing achieving a virtually complete removal of the initial CIP concentration within 60 min (Figure S1b).

Afterwards, we tested the effect of different CIP concentrations, pH level, and catalyst dosage on the efficiency of the catalytic process (

Table 1. k and %DR data regarding the optimization of the CIP concentration, pH level and catalyst dosage of the photodegradation of CIP with NC_50:50. The corresponding plots are found in Figure S1.

Test	Parameter Value	k (min−1)	%DR	Time (min)
CIP concentration	10 ppm	0.0823	96.4	40
	20 ppm	0.0537	96.2	60
	30 ppm	0.0200	69.7	60
pH level	5	0.0256	78.5	60
	7	0.0828	99.1	60
	9	0.0765	98.8	60
Catalyst dosage	10 mg	0.0218	73.4	60
	20 mg	0.0763	99.1	60
	40 mg	0.1114	98.0	40
	60 mg	0.2372	98.2	16

and Figure S1c–h). Starting with the initial CIP concentration, we found, as expected, that the lowest CIP concentration led to the highest reaction rate constant (0.0823 min⁻¹ for 10 ppm versus 0.0537 and 0.0200 min⁻¹ for 20 and 30 ppm, respectively). Although using 10 ppm of CP resulted in a higher rate constant, 20 ppm was selected for subsequent assays because it still achieved near-complete degradation using a higher pollutant concentration within a relatively short time frame.

Regarding the pH level, and already using a CIP concentration of 20 ppm, the results revealed that a pH of 7 resulted in the highest catalytic efficiency, achieving a rate constant of 0.0828 min⁻¹ and a %DR of 99.1% within 60 min of irradiation (Figure S1e,f). Finally, regarding the catalyst dosage, using 20 ppm of CIP at pH 7, we found that a 60 mg dosage led to the best results, resulting in a rate constant over double that of second-best performer (0.2372 min⁻¹ versus 0.1114 min⁻¹ for a 40 mg dosage) and achieving a virtually complete degradation of the initial pollutant within 16 min of irradiation (Figure S1g,h).

While this dosage was the highest amount we tested, we decided not to continue the test further, as higher concentrations resulted in a large amount of catalyst being deposited at the bottom of the reaction mixture. Thus, there was no advantage in adding more material. In summary, for subsequent tests, we selected the following reaction conditions: 20 ppm of CIP, pH 7, and 60 mg of catalyst, with a total reaction volume of 100 mL.

Based on these reaction parameters, the photocatalytic efficiency of NC_50:50 was compared to that of commercial TiO₂, photolysis, and CDs by themselves (Figure 1a,b). Under the same conditions, NC_50:50 exhibited a rate constant of 0.2372  ±  0.0032 min⁻¹ and a %DR of 98.2 ± 0.6% within 16 min of irradiation, outperforming commercial TiO_2, which achieved a rate constant of 0.0749  ±  0.0002 min⁻¹ and a %DR of 89.2  ±  1.7% after 30 min under simulated sunlight. Furthermore, the effect of photolysis was minimal, with only 5.8% degradation observed after 30 min. Photodegradation experiments using CDs alone yielded comparable results to photolysis, indicating that CDs by themselves are not sufficient to degrade CIP under the same conditions. These findings demonstrate that incorporating the CDs into TiO₂ substantially enhances the semiconductor’s photocatalytic performance under solar irradiation.

By analyzing the evolution of the chromatogram (Figure 1c) with increasing reaction times, we can observe that CIP’s initial peak, found at around retention time (RT) of 3.65 min, disappears quite rapidly (within 16 min of irradiation). Several peaks ascribed to potential degradation by-products appear during the course of the process. However, they also disappear as the photocatalytic process continues. Namely, when we prolonged the duration of the photocatalytic assay (Figure 1d), we observed that no significant peaks (with over 1% of initial CIP peak area) are found for an irradiation time of 300 min, suggesting that even the produced by-products are also degraded by NC_50:50. To better demonstrate this, a full overview of the evolution of chromatogram peaks areas appearing over the course of the degradation and their evolution over time can be found in Figure 1e (with specific data on Table S1), which clearly demonstrated that all additional peaks appearing over the course of the experiment (P2–P6) tend to disappear when compared to the initial CIP peak (P1).

To further assess the impact of incorporating CDs into TiO₂ on the catalytic performance, we quantified the enhancement of TiO₂’s catalytic efficiency upon the incorporation of CDs by using Equation (3):

$$\%\ln \left(\mathrm{C}\right)=100 \times {\displaystyle \frac{k_{NC}-k_{Ti{O}_2}}{k_{Ti{O}_2}}}$$

(3)

In this context, $\%\ln \left(\mathrm{C}\right)$ serves as a comparative metric for evaluating the enhancement of TiO₂’s catalytic efficiency upon the incorporation of CDs regarding the photodegradation of CIP, $k_{NC}$ represents the reaction rate constant of NC_50:50, while $k_{Ti{O}_2}$ corresponds to the reaction rate constant of commercial TiO₂. By applying this formula, we estimated the % ln(C) to be 217%, demonstrating a very significant enhancement in catalytic performance compared to that of commercial TiO₂. This was achieved by adding CDs, which, in addition to improving the catalytic performance, utilized corn stover in a manner that upcycled this otherwise discarded material.

However, aside from the increased photocatalytic efficiency, it is also worth noting that adding CDs also increased the adsorption of CIP to NC_50:50 in the dark compared to TiO₂ (Figure S2a). In fact, nearly half of the initial CIP concentration seems to be adsorbed by NC_50:50 during the 30 min period in the dark. This is relevantly higher than what is observed for TiO₂, where only around 4.1% of the initial CIP appears to be adsorbed. To assess if this higher adsorption could be responsible for the observed photodegradation with the lamp on, we compared the variation in C_t/C₀ of CIP in the presence of NC_50:50 both with the lamp on and off (Figure S2b). These results indicate that while there is a significant adsorption of CIP into the _NC50:50 during the 30 min dark period before turning on the lamp, the adsorption is stabilized thereafter. That is, after the 30 min dark period, when the lamp is still off, the C_t/C₀ ratio remains constant. While the lamp is on, a significant decrease in CIP levels is observed during the same period. These results confirm that the degradation observed during photocatalytic assays is due to photocatalysis itself, and not just continuous adsorption into NC_50:50.

The different degrees of adsorption between NC_50:50 and TiO₂ can be explained by the fact that by adding a nanoparticle with a highly functionalized surface [31] with different functional groups that confer a better affinity towards organic compounds than that inherent to TiO₂, the photocatalyst facilitates the adsorption of CIP, enhancing its interaction with active sites.

In fact, data from BET analysis revealed that NC_50:50 presents a higher specific area (S_BET) than TiO₂ (

Table 2. Surface area, pore volume and pore size for TiO₂ and NC_50:50.

Sample	SBET (m2 g−1)	Vtotal (cm3 g−1)	Dp (nm)
TiO2	105.2	0.276	13.1
NC50:50	269.9	0.212	4.1

). Namely, the results suggest that both TiO₂ and NC_50:50 exhibit Type III isotherms (Figure 2a), with the nanocomposite showing an SBET of 269.9 m²g⁻¹, over 2.5× higher than that of TiO₂ (105.2 m² g⁻¹). This increased surface area helps to explain the higher CIP adsorption observed with NC_50:50. Regarding pore size, both TiO₂ and NC_50:50 are classified as mesoporous, presenting pore widths of 13.1 and 4.1 nm, respectively (estimated via the BJH method with pore size distribution represented in Figure 2b) [52].

2.3. Scavenger Assays

To investigate the reactive species responsible for the enhanced catalytic activity of NC_50:50, sunlight-driven CIP degradation assays were conducted in the presence of specific radical scavengers. Hydroxyl radicals (·OH), superoxide radicals (O₂⁻), electrons (e⁻), holes (h⁺), and singlet oxygen (¹O₂) were selectively scavenged using isopropyl alcohol (IPA, 10 mM), 1,2-dihydroxybenzene-3,5-disulfonate (Tiron, 10 mM), potassium nitrate (KNO₃, 10 mM), potassium iodide (KI, 10 mM), and sodium azide (1 mM), respectively. These scavengers were selected based on their established roles in targeting reactive species commonly associated with TiO₂ and TiO₂-composites [41]. IPA and Tiron were of particular interest, as hydroxyl and superoxide radicals, their respective target species, are typically among the primary reactive species in the photocatalytic mechanism of bare TiO₂ [26,53].

The influence of each reactive species was assessed by measuring the change in the %DR in the presence of the scavengers, as calculated using Equation (4) [15]:

$$\%DRV=\%{DR}_{test}-\%{DR}_{control}$$

(4)

In this equation, %DRV represents the variation in %DR when a scavenger is introduced (%DR_test) relative to the control test (%DR_control), which represents standard photodegradation without scavengers. Therefore, the magnitude of the %DRV allows for a quantitative assessment of each species’ contribution: a higher degree of inhibition (more negative %DRV) corresponds to a greater relative importance of that species in the degradation process.

The results from the scavengers’ tests revealed that, for NC_50:50, all tested reactive species, except for the hydroxyl radical, participated in the photodegradation mechanism (Figure 3a). Namely, tests using IPA did not yield any significant %DRV value. For their turn, the %DR was diminished significantly in the presence of Tiron (−51.1 ± 5.4%), KI (−40.9 ± 3.5%), KNO₃ (−22.3 ± 0.8%) and sodium azide (−19.0 ± 1.0%). This indicates that the superoxide radical, h⁺, e⁻ and ¹O₂, by that order, have a role in the photocatalytic mechanism of NC_50:50. Regarding TiO₂ (Figure 3b), the reactive species responsible for the degradation mechanism were the superoxide radical, h^+, and ¹O₂.

2.4. Recycling Studies

Recycling experiments for the photodegradation of CIP under simulated solar irradiation were conducted to evaluate the stability and performance of NC_50:50 over three consecutive runs. In this experiment, consecutive cycles of CIP photodegradation were performed in 1 h periods of solar irradiation in the presence of NC_50:50. After each cycle, the composite was collected, dried, and reused repeatedly without further processing.

It was observed that the %DR obtained in the final cycle was still over 95%, compared to 99% in the first cycle (Figure 4). Given this, NC_50:50 is considered to have good performance stability and still achieves a nearly complete degradation of the target pollutant within the same time period. The number of considered cycles is also consistent with literature data [54,55,56].

It should be noted that there were mass losses during material recovery that made it difficult to perform additional cycles. For one, this issue is nevertheless common for powder-based photocatalysts. In that sense, it can be considered a general limitation for this type of photocatalyst, and further exploration should be the focus of future studies. Namely, the addition of the nanocomposite to lightweight materials to provide floating capability, which should allow for facile separation from the matrix to be treated by skimming. Nevertheless, the obtained %DR variation over the performed cycles still allows us to characterize the photocatalytic activity of the material as stable.

2.5. Comparison to the Literature

To better evaluate the effectiveness of our composite, the photocatalytic performance of NC_50:50 was compared to other TiO₂-based systems reported for the photodegradation of CIP. The comparison (Table 3) was made regarding k and how much the k obtained by NC_50:50 is enhanced relative to literature entries (% In(C)).

First, NC_50:50 presents the highest rate constant (0.2372 min⁻¹). In fact, the nanocomposite studied in this work presented a k that was enhanced, on average, by 1444% over all the other entries. Namely, the observed enhancement was over 500% for all entries (reaching enhancement values of 2691% and 5303%), except for entries 3 and 7. However, even considering those latter entries, the enhancement was also quite relevant (133% and 84%, respectively).

Furthermore, it should be noted that entry 3 only considered UV radiation (in which TiO₂ is known to be intrinsically more efficient), and not sunlight (as the present work did). Regarding entry 7, the measured k was obtained by considering about double the catalyst dosage and half of the CIP concentration in the present study. Finally, while entry 7 did employ artificial solar radiation as an irradiation source, contrary to our study, the performance of the photocatalyst was not validated under natural sunlight (Section 2.6).

Moreover, it is worth noting that this was achieved employing CDs partially prepared from corn stover. Meaning that by using our waste-incorporating carbon nanoparticles, we not only obtain a significantly higher catalytic potential than other alternatives, but we also use otherwise discarded organic material, thus providing a way to upcycle this organic waste.

Future work should include the evaluation of NC_50:50’s performance against other antibiotics, such as tetracycline and oxytetracycline [57,58,59,60,61,62].

Table 3. Comparison between the rate constant (k, min⁻¹) obtained by NC_50:50 and other TiO₂-based photocatalysts for the photodegradation of CIP. Here, % ln(C) is also calculated according to Equation (3), but considering the k of each literature entry as a reference. Namely, here, % ln(C) measures how enhanced the k of NC_50:50 is over each literature entry.

Photocatalyst	Cat. (g L−1)	CIP Conc. (ppm)	Lamp	k (min−1)	% In(C)	DOI
TiO2/CDs NC50:50	0.6	20	Solar simulator	0.2372		This work
Glass-deposited TiO2	0.13	20	365 nm 10 mW/cm2	0.0085	2691	[63]
Graphitized carbon-TiO2	0.35	15	254 nm 14 W	0.102	133	[64]
Glass-deposited TiO2	1	3	UV	0.0197	1104	[65]
TiO2 NPs	0.7	80	3 × 254 nm 8 W	0.00439	5303	[66]
TiO2/SnO2	1	2.5	3 × 253 nm 35 W	0.0311	663	[67]
TiO2/CDs6 wt%	1	10	Xe Arc 350 W 290 Filter	0.129	84	[68]
TiO2/H2O2	0.5	9	Solar 800 W Xe lamp	0.022	978	[69]
TiO2/WO3	0.5	20	Sunlight	0.034	598	[70]

Table 3. Comparison between the rate constant (k, min⁻¹) obtained by NC_50:50 and other TiO₂-based photocatalysts for the photodegradation of CIP. Here, % ln(C) is also calculated according to Equation (3), but considering the k of each literature entry as a reference. Namely, here, % ln(C) measures how enhanced the k of NC_50:50 is over each literature entry.

Photocatalyst	Cat. (g L−1)	CIP Conc. (ppm)	Lamp	k (min−1)	% In(C)	DOI
TiO2/CDs NC50:50	0.6	20	Solar simulator	0.2372		This work
Glass-deposited TiO2	0.13	20	365 nm 10 mW/cm2	0.0085	2691	[63]
Graphitized carbon-TiO2	0.35	15	254 nm 14 W	0.102	133	[64]
Glass-deposited TiO2	1	3	UV	0.0197	1104	[65]
TiO2 NPs	0.7	80	3 × 254 nm 8 W	0.00439	5303	[66]
TiO2/SnO2	1	2.5	3 × 253 nm 35 W	0.0311	663	[67]
TiO2/CDs6 wt%	1	10	Xe Arc 350 W 290 Filter	0.129	84	[68]
TiO2/H2O2	0.5	9	Solar 800 W Xe lamp	0.022	978	[69]
TiO2/WO3	0.5	20	Sunlight	0.034	598	[70]

2.6. Photocatalysis Under Natural Sunlight

Finally, the last stage of our experiment consisted of analyzing the photocatalytic potential of NC_50:50 under natural sunlight to further assess its practical applicability. Namely, an experiment was conducted on the roof of the FC3 Building of the Faculty of Sciences—University of Porto (Portugal), in which the photodegradation of CIP under direct sunlight irradiation was monitored (Figure 5 and

Table 4. Prevailing atmospheric conditions and %DR obtained concerning the photocatalytic degradation of CIP by NC_50:50 under direct sunlight. Meteorological data obtained from Instituto Português do Mar e da Atmosfera (IPMA).

Day	Time Period	Temperature Range	UV Index	Conditions	%DR	Time to ~100%DR
23 August 2024	11.00 to 15.00	30–32 °C	8	Sunny	~100%	30 min
24 August 2024	11.00 to 15.00	26–29 °C	8	Clouded	~100%	30 min
25 August 2024	11.00 to 15.00	23–26 °C	9	Clouded	~100%	15 min

). The assays were performed during three consecutive days with different meteorological conditions (Table 4).

The results demonstrate that, even under real sunlight, NC_50:50 is very efficient at degrading CIP. In fact, on the day experiencing the highest UV index (Day 3), NC_50:50 achieved an estimated reaction rate constant of 0.2676 min⁻¹ (Figure 5a), which was even higher than that obtained with the solar simulator as the irradiation source (0.2372 min⁻¹). The photocatalytic potential was further confirmed over three different days, in which CIP was virtually fully degraded within 30 min of irradiation, with Day 3 recording a %DR of ~100% within 15 min of exposure to sunlight (Figure 5b).

Interestingly, neither the temperature nor the atmospheric conditions seem to affect catalytic efficiency significantly. Instead, photocatalytic activity appears to be more closely related to variations in the UV index, as increasing the index from 8 to 9 led to halving the irradiation time required to achieve a %DR of 100%. It should be noted that the UV index is a meteorological measure of the intensity of sunburn-producing UV radiation, which is used here as a proxy for the intensity of natural sunlight (which includes visible light). The enhanced activity of the photocatalyst is attributed to its higher visible light absorption, which allows efficient utilization under sunlight.

In terms of degradation products, as observed in Figure 5c–e, the same peaks (with the same retention times) appear in the chromatograms in the same order as they did when using the solar simulator, suggesting that the degradation products are the same, independently of employing simulated solar irradiation or real sunlight. It should be noted that these potential degradation by-products were also degraded by NC_50:50 under natural sunlight, similar to what was observed under simulated sunlight.

The ability of NC_50:50 to degrade CIP under natural sunlight in an efficient manner can be attributed to the addition of CDs to TiO₂. Namely, CDs are known to be good electron acceptors during photoinduced electron transfer, leading to an improved separation of photoexcited e⁻/h⁺ pairs [43]. In fact, previously performed PL and EIS assays involved NC_50:50 (mentioned in Section 3.1) [16] indicated that the addition of CDs led to suppression of the recombination of TiO₂’s photoinduced charge carriers and to lower photoinduced electron transfer resistance. The NC_50:50 was also previously shown to possess enhanced visible light absorption than TiO₂ [16], which can be a result of the typical broadband absorption range of CDs [15,44,45]. Given that the main individual component of sunlight is visible light, this enhancement in visible light absorption indicates a more efficient activation under visible light. Herein, the superior performance of NC_50:50 under real-world irradiation conditions can be explained by the addition of CDs, which enhances sunlight-driven photoactivation and facilitates efficient charge separation, ultimately improving the photonic efficiency of the photocatalyst.

2.7. Photocatalytic Mechanism

Scavenger assays indicated that superoxide radical, h⁺, e⁻ and ¹O₂ (in decreasing order of magnitude) all had a relevant role in the photodegradation of CIP (Figure 3a), when catalyzed by NC_50:50.

It should be noted that while TiO₂-based photocatalysts are mainly associated with the production of hydroxyl radicals (under UV radiation) [71], their reactive species generation profile can vary at different irradiation wavelengths. Namely, different groups observed that superoxide anion was the primary reactive oxygen species (ROS) in the visible light-driven activity of TiO₂-based photocatalysts [72,73]. Also under visible light, Zhou et al. found that ¹O₂ to be a main reactive species in the studied TiO₂-based system [74]. By their turn, Guo et al. attributed the sunlight-driven degradation of organic dyes by TiO₂ photocatalysts to hole-mediated oxidation, rather than to hydroxyl radicals [75]. Thus, our results are consistent with existing literature.

As for the specific photocatalytic mechanism, as mentioned in Section 2.1 [16], the addition of CDs to NC_50:50 appeared to have led to an inhibited recombination of photogenerated charge carriers. This is attributed to the capacity of CDs (among other carbon nanomaterials) to act as electron reservoirs, thereby retaining photoexcited electrons transferred from TiO₂ upon light irradiation [15,43,76]. This enhances charge separation and suppresses the recombination of holes and electrons [43,77]. Therefore, effectively extending their lifetime and allowing them to have a more direct role in the photodegradation reaction. This can explain the more prominent role of both h⁺ and e⁻ in the photodegradation of CIP observed in the study (Figure 3a).

The direct involvement of h⁺ and e⁻ might also be facilitated by the enhanced adsorption capability of NC_50:50 (Figure 2 and Figure S2, Table 2). Namely, it was observed that the addition of CDs led to a 2.5× increase in the specific surface area of the nanocomposite, which resulted in a large adsorption of CIP to NC_50:50 (Figure S2), during the dark period. This means that before the start of light irradiation, CIP molecules are already positioned at the surface of the photocatalyst, where they can interact more rapidly with photogenerated h⁺ and e⁻.

Furthermore, CDs are typically considered both photoinduced electron acceptors (which allows them to retain e⁻ from TiO₂) and donors [78]. The latter characteristic can also enable them to act as mediators/electron relays [79,80], and assist in the formation of superoxide anion. This latter ROS can then either induce the degradation of CIP or be oxidized in a hole-mediated process, to generate ¹O₂ [81]. The efficiency of these processes is expected to be enhanced by the inhibited recombination rate of photogenerated charge carriers, which justifies the observed involvement of superoxide anion and ¹O₂ in the degradation of CIP (Figure 3a).

In short (Figure 6), the addition of CDs allows for sunlight-driven photoexcitation and enhanced charge separation, by retaining photoexcited electrons from TiO₂. Given their extended lifetime, these e⁻ can generate superoxide anion, which is partially converted to ¹O₂ via a h⁺-mediated route. These two ROS, together with remaining h⁺ and e⁻, directly oxidize CIP. The degradation process is also expected to be enhanced by the larger surface area of NC_50:50, resulting from the addition of CDs, which allows for significant CIP adsorption prior to the irradiation phase. This pre-positions CIP in the reactive interface, which can enhance reaction rates once light irradiation starts.

It should also be highlighted that the addition of CDs also modulates the reactive species generation profile of the resulting photocatalytic nanocomposite. Namely, analysis of Figure 3 shows that the reactive species involved in CIP degradation differ for reactions catalyzed by NC_50:50 or commercial TiO₂. As mentioned, superoxide anion, h⁺, e⁻ and ¹O₂ are involved in photodegradation catalyzed by NC_50:50. However, only superoxide anion, h⁺, and ¹O₂ were found to be involved in the commercial TiO₂-catalyzed process.

Furthermore, the difference between reactive species profiles does not appear to end with the involvement or not of e⁻. That is, the NC_50:50 appears to be more balanced in terms of the reactive species involved (Figure 3). While higher, the impact of superoxide anion is close to that of h⁺. The impact of e⁻ is similar to that of ¹O₂, and their combined contribution to CIP degradation is still significant (about half of the combined contributions of superoxide anion and h⁺).

The profile is different for commercial TiO₂ (Figure 3). Now, superoxide anion is more clearly the main contributor by a relevant margin to h⁺, and that of ¹O₂ is relatively residual. These differences can be ascribed to the charge separation-enhancing effect of CDs. Thereby, allowing the higher involvement of addition of h⁺ and e⁻, and of additional ROS (such as ¹O₂).

Thus, the incorporation of CDs not only enhances photocatalytic efficiency but also appears to alter the charge-transfer dynamics of TiO₂. By modulating the reactive species generation profile, CDs can change the available degradation pathways. This finding potentially opens a pathway for the rational design of sunlight-active TiO₂-based photocatalysts with tunable surface reactivity and broader contaminant compatibility.

3. Materials and Methods

3.1. Preparation of CDs

The CDs were synthesized via a hydrothermal method, reproducing a synthesis procedure previously developed by us and thus, the morphological, structural, surface, and luminescent characterization of the material can be found in [16,82]. In summary, citric acid (CA), corn stover, and ethylenediamine (EDA) (1:1:2 mass ratio, in a total of 0.3 g) were dispersed in a NaOH solution (0.01 M). The reaction mixture was placed in a Teflon-lined autoclave reactor, which was then sealed stainless-steel outer shell. The mixture was subsequently subjected to a hydrothermal treatment (200 °C for 8 h), after which it was left to cool overnight. The resulting mixture was then subjected to purification by centri-fugation (6000 rpm for 30 min, while retaining the supernatant), and dialysis (3.5 kDa cutoff, 24 h). The former was performed to remove insoluble material, while the latter was performed to eliminate any remaining soluble contaminants that could interfere with the physicochemical properties [33]. Finally, the purified CD solution was lyophilized, weighed, and then redissolved in deionized water before being stored in the refrigerator (4 °C) until needed.

3.2. TiO₂/CDs Nanocomposites Preparation

To synthesize the TiO₂-CD nanocomposite, we modified a procedure previously developed by us [83,84], adapting it to include CDs [16]. In brief, titanium isopropoxide (TTIP) was initially added to IPA dropwise under continuous stirring. Afterwards, deionized water was introduced to precipitate TiO₂. A controlled amount of CDs, equivalent to 4% of the total expected mass, was then added dropwise under stirring. Before incorporation, the CD solution was maintained under stirring at 70 °C to ensure homogeneity. After adding the CDs, the suspension was stirred for an additional 2 h before filtration, followed by drying overnight and annealing at 300 °C in a muffle furnace (Thermolyne 47900, 1000 W, Thermo Scientific, Waltham, MA, USA, EUA), resulting in NC_50:50. In this nomenclature, 50:50 indicates that corn stover and CA were used in a 50:50 mass ratio as co-carbon precursors in the synthesis of CDs, which are then used in the preparation of the nanocomposite. NC_50:50 was then kept in the dark until further use.

3.3. Photocatalytic Activity Assays

The sunlight-driven photocatalytic efficiency of commercial TiO₂ and NC_50:50 was evaluated using an Ossila LED Solar Simulator (1000 W m⁻², Ossila, Sheffield, UK) as the irradiation source for the photodegradation of CIP. For these assays, 60 mg of photocatalyst was dispersed in 100 mL of CIP solution (20 mg L⁻¹). Afterwards, the mixture was exposed to 1 min of ultrasonic agitation. Prior to switching on the irradiation source, the suspension was stirred in the dark for 30 min to reach an adsorption–desorption equilibrium [85]. Samples were collected at the beginning of that dark period (t = −30 min), to assess pollutant adsorption on the catalyst). Following this period, the solar simulator was turned on, and samples were collected at specific time intervals: at the start of irradiation (t = 0 min), and at several different irradiation times. To remove the catalyst from the sample prior to analysis, the aliquots were centrifuged at 10,000 rpm for 10 min and filtered using PTFE 0.45 µm syringe filters (LLG Labware).

The chromatographic analysis of the samples was performed using a reverse-phase HPLC with a diode array detector (RP HPLC-DAD) chromatographic system. The system employed a Thermo Scientific SpectraSystem P1000 pump (Thermo Scientific, Waltham, MA, USA, EUA), a Rheodyne manual injection valve, an Acclaim™ 120 column and a UV6000 LP diode array detector from Thermo Finnigan (San Jose, CA, USA). A mobile phase composed of 49.5% methanol, 49.5% deionized water, 1% acetic acid, and 1 mM ammonium acetate was used with a flow of 0.65 mL per minute.

3.4. Scavenger Assays

To gain insight into the potential participation of different reactive species in the photocatalytic degradation of CIP, additional photocatalytic assays were conducted using scavengers for each respective species, and the effect of the scavengers was monitored via RP-HPLC-DAD. Herein, IPA (10 mM) [26], Tiron (10 mM) [86], KNO₃ (10 mM) [87], KI (10 mM) [26], and sodium azide (1 mM) [88] were used to evaluate the roles of hydroxyl radicals (·OH), superoxide radicals (O₂⁻), electrons (e⁻), holes (h⁺) and singlet oxygen (¹O₂), respectively.

3.5. BET Analysis

The surface area and pore parameters were determined using a nitrogen adsorption analyzer (TriStar Plus, Micromeritics, Norcross, GA, USA). Prior to conducting the N₂ adsorption–desorption measurements, the samples were degassed under N₂ flow at 200 °C for 6 h to remove adsorbed water and volatile impurities, ensuring that the measured surface area and pore characteristics reflect only the properties of the materials and not those of surface contaminants. The pore size distributions were calculated using the desorption branch of the N₂ isotherms, applying the Barrett–Joyner–Halenda (BJH) method.

4. Conclusions

This work demonstrates the efficient sunlight-driven degradation of the pharmaceutical contaminant CIP using a TiO₂/CDs nanocomposite under real-world irradiation conditions, a crucial step toward large-scale water-treatment applications. Furthermore, the used CDs were prepared from a major agricultural waste, corn stover.

Initial testing was performed under simulated sunlight. Under these conditions, NC_50:50 achieved a virtually full degradation of CIP within 20 min, with a rate constant of 0.2372 min⁻¹. Moreover, the photocatalytic activity of the nanocomposite was found to be 217% enhanced compared with commercial TiO₂. NC_50:50 also showed good stability. Finally, while some potential by-products were detected, they were also photodegraded by NC_50:50. When compared with other photocatalysts used for CIP degradation, NC_50:50 showed the most promising performance.

Subsequently, the performance of NC_50:50 was validated under real-world irradiation conditions. Namely, under natural sunlight over three consecutive days, with different meteorological conditions. This photocatalyst was found to retain its efficiency, fully photodegrading CIP under 30 min, irrespective of the day. Potential degradation by-products were also formed but were subsequently degraded by NC_50:50. This indicates that the nanocomposite exhibits enhanced photodegradation catalytic effects even under real-world irradiation conditions.

The enhanced activity of NC_50:50 was attributed to a combination of effects that resulted from the incorporation of CDs. More specifically, their incorporation resulted in more efficient activation by sunlight, due to enhanced visible-light absorption. It was also concluded that CDs suppressed the recombination of holes and electrons, increasing the photodegradation efficiency. The incorporation of CDs also increased the specific surface area of the photocatalyst, which enhanced CIP adsorption before the photocatalysis stage. This pre-positioned CIP in the surface interface, which facilitated the photodegradation after starting the light irradiation phase.

Finally, the incorporation of CDs not only significantly improved the sunlight-driven photocatalytic activity but also modulated the reactive species generation profile of TiO₂. By altering the type and contribution magnitude of generated reactive species, CDs can modify the available degradation pathways. This finding potentially opens a pathway for the rational design of sunlight-active TiO₂-based photocatalysts with tunable surface reactivity and broader contaminant compatibility.