Evaluation of UV/TiO₂/H₂O₂ photocatalysis for the removal of perfluorinated organic compounds from water

Abstract

Conventional processes do not eliminate PFAS in wastewater treatment plants, so they are found in trace concentrations in wastewater and rivers. Perfluorinated organic pollutants (PFAs) are toxic and persistent, that have a significant risk to human health and ecosystems. Perfluorooctanoic acid (PFOA) is one of the most common PFAs to enter water bodies, and does so primarily through untreated wastewater. Advanced oxidation processes (AOPs), in which hydroxyl free radicals •OH are generated, have been able to degrade persistent pollutants. This study aimed to evaluate the efficiency of the UV/TiO₂/H₂O₂ system, an AOP, in degrading PFOA in both deionized and surface water. A 1000 mL cylindrical glass photoreactor equipped with a cooling jacket, a cylindrical quartz tube, and 254 nm and 360 nm UV lamps was used. In UV/TiO₂/H₂O degradation tests, TiO₂ concentrations ranging from 20 to 1500 mg/L, H₂O₂ concentrations ranging from 2.22 to 6.66 g/L, and PFOA concentrations ranging from 50 to 400 µg/L were evaluated at pH 7 and 300 min of UV irradiation. A degradation of approximately 26.08% was achieved after 5 h of treatment in deionized water. Optimal UV/TiO₂/H₂O process conditions in surface water showed a degradation of 20.09%. The PFOA photodegradation experiments using TiO₂ were not straightforward, due to the high persistence of PFOA and the fact that PFOA adsorbs onto glass, making chromatographic analysis of PFOA difficult. This was overcome by rigorously monitoring the traceability of PFOA adsorption on glass to ensure that the amount of PFOA degradation with TiO₂ was accurately measured.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-34613-z.

Introduction

PFAS are divided into two main categories: perfluoroalkyl and polyfluoroalkyl substances. The former are molecules with carbon chains that are completely fluorinated, except for the terminal end, which has a polar functional group, such as carboxylates or sulfonates¹. These are subdivided into perfluoroalkyl carboxylates (PFCAs), such as PFOA, and perfluoroalkyl sulfonates (PFSAs), such as PFOS¹. Polyfluoroalkyl compounds have an aliphatic chain that is not completely fluorinated, meaning they have at least one carbon atom with fluorine, hydrogen, or oxygen atoms. PFOA is a polar and highly persistent compound². The physical and chemical properties of PFOA are presented in Table 1.

Table 1.

Physical and chemical properties of PFOA.

PROPERTY	PFOA
Molecular structure
Molecular formula	C8HF15O2
Appearance at ambient T* and P**	Waxy white solid
Molecular mass	414 g/mol
Vapor pressure	10mmHg (25 °C)
Solubility in pure water	9.5 g/L (25 °C)
Melting point	45–50 °C
Boiling point	189–192 °C (736 mmHg)
pKa****	2.5
Half-life in water	92 years (25 ° C)
Henry’s constant	Not measurable
Koc ****	2.06

*Temperature, **Pressure, ***Acid dissociation constant, ****Organic carbon partition coefficient.

PFOA is used in the manufacture of various products^3,4, including kitchen utensils, beauty and personal care products, cleaning products, firefighting foams, floor polishes, lubricants, and items such as upholstery, carpets, paints, surfactants, and emulsifiers, among others^1,5,6.

The production of PFAS is restricted by the Stockholm Convention (2009). In 2024, the Environmental Protection Agency (EPA) created the National Primary Drinking Water Regulation, which established maximum levels (MCLs) for six contaminants, including perfluorooctanoic acid (PFOA), for which it set an MCL of 4.0 ppt⁷.

Conventional processes do not remove PFOA in wastewater treatment, with the result that its presence in aquatic environments has been confirmed^6,8,9. PFOA concentrations between 2 and 70 ppb and between 0.1 and 400 ng/L⁷have been found in drinking water. They have also been found in surface water, air, soil, and food^10–13.

PFOA can cause liver damage¹⁴ and cell apoptosis in the eye, heart, and brain¹⁵ and induce DNA damage¹⁶ in zebrafish embryos and adults. These compounds can interfere with human and animal metabolism^17,18 are carcinogenic¹⁰, can affect newborns,¹⁹, and cause other human and animal health problems^20–23.

Therefore, treatment alternatives should be explored to remove PFOA in water treatment plants. Granular activated carbon adsorption, reverse osmosis, ion exchange resin adsorption, and AOPs, among others, have been used to remove PFOA from water^24,25. Adsorption has been shown to be effective, although it requires long treatment times and does not remove contaminants, but rather transfers them to the solid phase. Sonochemical and electrochemical processes have been used to remove PFOA, but they have the disadvantages of high energy consumption and low defluorination efficiency^26–30.

AOPs have been able to degrade some PFAS^31–33. AOPs oxidize contaminants rather than accumulating them in the solid phase, as occurs with adsorption. AOPs are much more efficient in a short period of time compared to traditional processes such as coagulation, flocculation, sedimentation, and filtration⁹. Several studies have been published on the treatment of PFAS and PFOA^14,34–37.

There are still gaps in the testing of alternatives that degrade and mineralize PFOA. Therefore, in this study, it was hypothesized that UV/TiO₂/H₂O₂ could degrade PFOA. The study evaluated the efficiency of the UV/TiO₂/H₂O₂ advanced oxidation process for PFOA degradation in aqueous matrices. Particular attention was paid to overcoming the analytical challenges caused by PFOA adsorption on glass surfaces, ensuring accurate degradation quantification.

Materials and methods

Materials

A 1000 mL cylindrical glass photoreactor (Fig. 1) was used; the photoreactor was completely covered with aluminum foil throughout all experiments. It was fitted with a cooling jacket (10 cm in diameter, including the water jacket, and 37 cm in height), and a cylindrical quartz tube (2.5 cm in diameter and 40 cm in height). 254 nm ultraviolet light lamp (8 W low-pressure mercury) and a 360 nm black light lamp (6 W) were used.

Fig. 1.

Execution of a photocatalytic experiment in the present work, which was always carried out covering the reactor with aluminum foil.

The following equipment was used: water pump; a magnetic stirrer plate; silicone hoses; polypropylene syringe; 1 μm glass fiber filters; polypropylene vials; polypropylene Falcon tubes.

The following substances were used: perfluorooctanoic acid (PFOA, C₈HF₁₅O₂) in solid state (95.0% purity) Merck (Germany); TiO₂ Degussa P-25; H₂O₂ at 30% w/v Panreac (Spain); anhydrous sodium thiosulfate (Na₂S₂O₃), purity > 99.0% from Merck (USA); and ultra-pure water obtained with a Milli-Q UV3R Direct system (18 MΩ cm electrical resistivity) (Millipore, France).

Experiments to optimize conditions for the photodegradation of PFOA

The solutions in the photoreactor were continuously stirred at 300 rpm using a magnetic stir bar. A quartz tube, containing an 8 W low-pressure mercury UV lamp emitting light at a wavelength of 254 nm, was placed inside the reactor. In some experiments, a 360 nm black light lamp was used. The temperature of the reaction solution was kept constant by circulating water around the reactor through a water jacket. The irradiation time was 5 h for all experiments. The experiments were performed in triplicate.

To determine the optimized reaction conditions, a series of initial experiments were performed to establish the concentration ranges of the reactants with which the reaction could be carried out. The experiments were conducted at an initial PFOA concentration of 200 µg/L. The surface-response methodology was used based on a central face-centered composite experimental design, including three central points. It should be noted that this methodology is useful in the modeling and analysis of problems in which a response of interest is influenced by several variables, and the objective is to optimize it³⁸. Thus, in the UV/TiO₂/H₂O₂ process, the effect of TiO₂ and H₂O₂ concentrations on the reaction was evaluated. Three concentration levels were considered for each variable. The results obtained were analyzed using Statgraphics Centurion XVI software (Statpoint, USA).

The experimental conditions for the PFOA removal tests using UV/TiO₂/H₂O₂ were: PFOA concentration: 200 µg/L, pH 7, reaction time (min): 300 min, H₂O₂ concentrations: 2.22, 4.44, and 6.66 g/L, TiO₂ concentrations: 100, 400, and 700 mg/L.

Sampling and filtration

The stopcock was purged, and the sample was extracted using a 5 mL syringe. A 0.22 μm pore-sized glass fiber filter was placed on the syringe tip to ensure the removal of TiO₂ particles. Subsequently, using syringe pressure, the sample was injected into a clean 15 mL Falcon flask, to which 700 µL of thiosulfate pentahydrate had been previously added. Once in the Falcon flask, a 300 µL volume of the sample was extracted using a micropipette. This volume was carefully transferred to a previously prepared 400 µL vial. 20 µL of 10% thiosulfate was added to the vial to stop the reaction and preserve the sample. Finally, the sample was stored refrigerated for later analysis. These samples were then filtered using a vacuum pump before being taken for chromatographic quantification.

PFOA removal under optimized conditions

To evaluate the reaction kinetics, experiments were conducted under different initial PFOA concentrations (400, 300, 200, 100, and 50 µg/L) but using optimized TiO₂ and H₂O₂ conditions.

Instrumental analysis and quality control

The determination of PFOA was by liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS). The system consisted of an ACQUITY UPLC H-Class chromatograph coupled to a XEVO TQD (Waters Corporation, United States). The analytical column used was a Kinetex C18 (2.1 × 100 mm, 2.6 μm) (Phenomenex, United States) and a PFAS Delay column (2.1 × 50 mm, 5.0 μm) (Restek Corporation, United States) was used prior to this as an injection port. The mobile phases consisted of (A) water: methanol (95:5 v/v) with 10 mM ammonium acetate and (B) methanol: water (95:5 v/v) with 10 mM ammonium acetate. The chromatographic method was isocratic 40:60 (mobile phase A: mobile phase B) at a flow rate of 0.3 mL min⁻¹ for 11 min. The column temperature was maintained at 30 °C and the injection volume was 10 µL. The retention time of PFOA was 4.24 min. The mass spectrometry was operated in negative electrospray ionization mode (ESI⁻) with multiple reaction monitoring (MRM) acquisition. Source and desolvation temperatures were maintained at 150 °C and 450 °C, respectively, with a capillary voltage of 3.0 kV. The quantification transition was m/z 413.0 → 369.0 (22 V cone voltage, 18 eV collision energy) and the confirmation transition was m/z 413.0 → 169.0 (22 V cone voltage, 10 eV collision energy).

The quantification of PFOA was done by external calibration. Calibration curves were prepared in water pH 2.0 in the range from 5 to 300 µg L⁻¹. The coefficients of determination were ≥ 0.995 and residuals in each calibration points were in the range ± 30% of the nominal concentration.

In each analytical batch (up to 20 samples) method blanks, solvent blanks, calibration verification standards (CCV), laboratory control samples (LCS) were included. Acceptance criteria were as follows: blank responses ≤ 1/3 of the reporting limit; CCV and LCS recoveries within ± 30% of target values. Routine verification of instrumental performance, background monitoring, and carry-over checks with methanol blanks were performed to ensure analytical integrity throughout the analysis.

Results and discussion

Optimization of H₂O₂ and TiO₂ concentrations for the photodegradation of PFOA

According to Fig. 2, in preliminary tests, increasing H₂O₂ to 4.44 g/L and TiO₂ to 20 mg/L increased the percentage of PFOA degradation. However, this percentage did not exceed 15%. While these low concentrations of TiO₂ have been shown to be effective in removing several persistent organic pollutants³⁹, their effectiveness in degrading PFOA is very poor^40–43. The rapid electron-hole pair recombination in TiO₂ limits the production of reactive oxygen species, which are essential for PFOA decomposition⁴⁴. Despite these limitations, other TiO₂ materials and modifications are still being actively investigated to improve its PFOA degradation capacity. However, these alternatives are more expensive, hampering scalability⁴⁵.

Fig. 2.

Preliminary tests with varying doses of TiO₂ and H₂O₂ (PFOA concentration of 200 µg/L; reaction time of 300 min, at a wavelength of 254 nm). (a) UV/10 mg/L TiO₂/1.11 g/L H₂O₂/200 µg/L PFOA; (b) UV/10 mg/L TiO₂/2.22 g/L H₂O₂/200 µg/L PFOA; (c) UV/20 mg/L TiO₂/2.22 g/L H₂O₂/200 µg/L PFOA; and (d) UV/20 mg/L TiO₂/4.44 g/L H₂O₂/200 µg/L PFOA.

The addition of H₂O₂ was expected to enhance PFOA degradation, as H₂O₂ can produce •OH upon activation by UV light. On the other hand, H₂O₂ can act as an electron acceptor, preventing the recombination of electrons and holes generated in TiO₂, maintaining catalytic activity^46,47. According to the literature, although H₂O₂ is a strong oxidant, its oxidation potential is not high enough to oxidize the C-F bonds of PFAS⁴⁸.

Since very low PFOA degradation rates were achieved at TiO₂ concentrations of 10 mg/L and 20 mg/L, higher concentrations were used in experiments involving PFOA degradation using TiO₂ photocatalysis^49,50. Meanwhile, H₂O₂ concentrations were double those obtained at the concentration with the best experimental results in the previous trials (4.44 g/L).

Table 2 shows the results obtained for the new TiO₂ concentration levels in triplicate experiments. An increase in the degradation obtained by increasing TiO₂ dosage is evident.

Table 2.

Results of PFOA degradation percentages using UV/TiO₂/H₂O₂.

PFOA (µg/L)	TiO2 (mg/L)	H2O2 (g/L)	Average	Standard Deviation
200	400	4.44	26.51%	4.96%
	20	8.88	13.89%	1.66%

254 nm UV light vs. 360 nm UV light

To improve the PFOA degradation obtained with 254 nm UV light, 360 nm UV light was used, so these experiments were performed with both light sources (254 and 360 nm). The results in Fig. 3 show that 254 nm UV light achieves greater PFOA degradation efficiency than 360 nm UV light, with higher percentages in four of the five TiO₂ concentrations evaluated, with the best results being at the lowest TiO₂ concentrations (100 and 400 mg/L). For this reason, it was decided to use the wavelength of 254 nm in the final experiments. TiO₂ photocatalysis using 254 nm UV light (UV-C) is generally more effective at degrading persistent organic compounds than 360 nm UV light (UV-A). While both wavelengths can induce photocatalysis, the shorter wavelength of 254 nm provides higher energy photons that are more readily absorbed by TiO₂, leading to a greater production of reactive oxygen species (ROS) and, consequently, enhanced degradation of pollutants.

Fig. 3.

Degradation of PFOA using UV-254 nm/TiO₂/H₂O₂ and UV-360 nm/TiO₂/H₂O₂ (PFOA concentration of 200 µg/L; H₂O₂ concentration of 4.44 g/L; and reaction time: 300 min).

Regarding 360 nm light, PFOA degradation increased at high TiO₂ concentrations, as observed in the treatment using 1500 mg/L of TiO₂, which obtained the highest PFOA degradation of approximately 28%. These results are consistent with what has been reported in the literature, that increasing TiO₂ concentrations using 360 nm light enhanced the degradation of the target contaminant⁵¹. An increase in PFOA degradation from 20% to 80% in 8 h was reported⁵⁰, when increasing the TiO₂ concentration, using 50 mg/L, 100 mg/L, 200 mg/L, 300 mg/L, 400 mg/L and 500 mg/L of TiO₂ and 254 nm light.

Central composite face-centered experimental design

At optimal levels of around 400 mg/L for TiO₂ and 4.44 g/L for H₂O₂, a central composite face-centered experimental design was defined around this region. Figure 4 presents the response surface obtained based on the experimental results, which indicates that PFOA removal, under the experimental ranges evaluated, is favored at TiO₂ and H₂O₂ concentrations at the intermediate levels used in this work.

Fig. 4.

Response surface for PFOA degradation using UV/TiO₂/H₂O₂ (initial PFOA concentration of 200 µg/L and reaction time: 300 min).

Statistical analysis

In the statistical analysis, the main effects plot shows that increasing the H₂O₂ concentration had a positive impact on PFOA degradation, reaching an optimum point around 4.44 g/L. The Pareto diagram shows that the quadratic effect of H₂O₂ has the most significant impact on PFOA degradation. However, it has a negative effect, with a nonlinear relationship between H₂O₂ concentration and process efficiency, which occurs at high H₂O₂ concentrations. The main effects plot shows that as the TiO₂ concentration increased, PFOA degradation also increased. However, there is a critical point when increased TiO₂ concentration no longer produces a proportional improvement in degradation, and the reaction efficiency begins to decline. The results showed that the optimized conditions for PFOA removal using photocatalysis with TiO₂ and H₂O₂, within the experimental ranges evaluated, are H₂O₂ concentration at 4.88 g/L and TiO₂ concentration at 605 mg/L.

Control tests

After defining the optimal reaction conditions, 5-h control tests were performed to evaluate PFOA degradation using different treatments (Fig. 5). All tests were conducted at a PFOA concentration of 200 µg/L and in triplicate. s tests were not performed, as s with 254 nm UV light has been reported not to degrade PFOA^52,53. s with 254 nm UV radiation might be insufficient for PFAS removal due to its chemical resistance, as this wavelength lacks the energy necessary to break the bonds of these substances^54,55. Similarly, oxidation with H₂O₂ in the dark shows very limited PFOA degradation. According to the literature, H₂O₂ alone is not very effective in the degradation of perfluorinated compounds, due to its chemical stability⁵⁶.

Fig. 5.

PFOA degradation using UV/TiO₂/H₂O₂ under optimized reaction conditions and control tests (initial PFOA concentration: 200 µg/L; H₂O₂ concentration: 4.88 g/L, TiO₂ concentration: 605 mg/L and isopropanol (scavenger) concentration: 200 mg/L).

The combination of UV and H₂O₂ slightly enhances PFOA degradation compared to H₂O₂ alone. Both H₂O₂ in the dark and TiO₂ without UV irradiation have very limited photocatalytic activity. In the absence of UV light, TiO₂ cannot generate the free radicals required for PFOA degradation. The combination of UV and TiO₂ enhances PFOA degradation. UV irradiation activates TiO₂, producing free radicals that can degrade PFOA more effectively than previous methods. These results are consistent with the findings of Quiroz-Vela et al. (2024), which showed that treatments that incorporated UV light exhibited superior degradation than those that did not⁵³. Sansotera et al. (2014), reported that TiO₂ (0.66 g/L) without UV radiation was not sufficient to produce PFOA degradation, and that there was no PFOA degradation with UV irradiation (75 W/m²) without a photocatalyst (s)⁵¹.

In order to clarify the role of hydroxyl free radicals in PFOA degradation, experiments were conducted in the presence of isopropanol, which is known to be a good.OH radical scavenger (reaction constant between the radical and isopropanol 1.9 × 109 mol L^− 1 s^− 1)⁵⁷. The presence of a scavenging agent such as isopropanol reduces the efficiency of PFOA degradation. Scavengers trap free radicals (such as •OH), reducing their availability to attack PFOA.

Effect of initial PFOA concentration

The effect of varying the initial PFOA concentration on its degradation was evaluated using UV/TiO₂/H₂O₂ under optimized conditions. The PFOA concentration varied between 50 and 400 µg/L. Figure 6 presents the results, indicating that PFOA degradation appears to decrease with increasing initial concentration of the compound.

Fig. 6.

Effect of initial contaminant concentration on PFOA UV/TiO₂/H₂O₂ degradation under optimized reaction conditions (H₂O₂ concentration: 4.88 g/L and TiO₂ concentration: 605 mg/L).

The reduction in photocatalytic efficiency can be attributed to the fact that the active sites of catalysts or activators such as TiO₂ can become saturated with an excess of organic molecules. This excess not only includes the main contaminant, but also the reaction byproducts⁶. Tang et al. (2006) reported that PFOA degradation during photocatalysis with TiO₂ and UV radiation decreased with increasing initial PFOA concentration⁵⁸. This was attributed to the saturation of the active sites on the photocatalyst surface and to the lower generation of hydroxyl radicals due to competition for adsorption sites on the catalyst. Sansotera et al. (2014), using a TiO₂ photocatalytic system at different initial PFOA concentrations (1656, 2898, 3312, and 4968 µg/L), found the highest percentage of degradation (32%) at the lowest PFOA concentration (1656 µg/L)⁵¹.

24-Hour experiment under optimized conditions

Figure 7 shows that 40% PFOA degradation was achieved after 24 h under optimized conditions. This indicates that PFOA is degrading very slowly. The reported times in the literature for PFOA degradation using UV/TiO₂ systems vary considerably. Chen et al. (2015) report a time of 12 h for four UV/TiO₂-based systems⁴⁹, while Li et al. (2016) report times of 7 hours⁵⁹, and Wang et al. (2021) a time of 3 hours⁵⁵.

Fig. 7.

24-hour experiment for PFOA degradation using UV/TiO₂/H₂O₂ under optimized reaction conditions (initial PFOA concentration of 200 µg/L; H₂O₂ concentration: 4.88 g/L and TiO₂ concentration: 605 mg/L).

Matrix effect: surface water

Surface water with the characteristics shown in Table 3 was used to perform PFOA photodegradation tests using the UV/TiO₂/H₂O₂ system under optimized conditions.

Table 3.

Characterization of surface water from a river located 80 km from the City of Medellín.

Parameter	Value
Chlorides (Cl-)	0.674 mg/L
Nitrates (NO3-)	0.223 mg/L
Sulfates (SO42-)	3.109 mg/L
Nitrites (NO2-)	< LCM (0.080 mg/L)
Fluorides (F-)	0.019 mg/L
Total Organic Carbon (TOC)	1.607 mg/L
pH	7.34
Apparent Color	25.53 UC
Turbidity	10 NTU

At 300 min, a PFOA removal rate of 20.09% was achieved in surface water (Fig. 8), slightly lower than that obtained in deionized water. Deionized water has much fewer ions than surface water, but it is precisely the large number of ions that promote the increased formation of free radicals such as nitrates. Nitrates can promote photodegradation because the s of nitrates (NO₃⁻) produces reactive nitrogen species (RNS), such as •NO₃ (∼2.3 V), •NO₂ (∼1.03 V), •NO₂ (∼0.39 V), and reactive oxygen species (ROS), such as •OH, ROO₂, and •O₂, which are potent active radicals for the degradation of organic compounds. Additionally, the surface water used for the experiment contains nitrates (Table 3). Thus, nitrates present in surface water did not significantly impact the photodegradation process, and it is possible that a scavenger present in the surface water influenced the decrease in PFOA photodegradation. Carbonates can scavenge free radicals⁶⁰, and surface water typically contains carbonates.

Fig. 8.

Comparison of PFOA degradation using UV/TiO₂/H₂O₂ under optimized reaction conditions in surface water and deionized water (initial PFOA concentration: 200 µg/L; H₂O₂ concentration: 4.88 g/L; and TiO₂ concentration: 605 mg/L).

The presence of inorganic ions such as chlorides, nitrates and sulfates in the treated matrix could be another factor associated with a reduction in process efficiency⁶¹. The negative influence of these species is related to the formation of iron complexes that decrease the availability of this factor in the system to the reaction with the free hydroxyl radical present, giving rise to other less reactive radicals (scavenger effect), or to the precipitation of these after reacting with iron ions dissolved in the solution^62,63. However, this effect will depend on the concentration of each ion in the system. In this sense, chloride and/or sulfate ions can trap the ferrous and ferric cations present, forming different types of complexes.

Furthermore, other organic compounds that are simultaneously present can compete for adsorption sites on the catalyst surface or oxidizing species, which can hamper the efficiency of photocatalysis, especially when these compounds are present at higher concentrations than the target contaminant^64,65.

Dissolved organic matter (DOM) is known to be a scavenger for •OH radicals, competing with the target pollutants. On the other hand, DOM can absorb radiation and compete with organic contaminant, reducing the generation of radicals⁶⁶.

Ali et al. (2024) reported that organic matter affected photocatalysis with rGO-nZVI catalyst in the presence of H₂O₂, thereby degrading PFOS and PFOA⁶⁷. They used organic matter concentrations of 3, 6, and 10 ppm total organic carbon (TOC), giving PFOS degradations 8%, 22%, and 21% lower, respectively, than in the absence of organic matter, and PFOA degradations 15%, 16%, and 33% lower, respectively, than in the absence of organic matter.

Giri et al. (2012) investigated the photodecomposition of PFOA in three types of water: ultrapure, tap, and surface water. The results showed that PFOA decomposition was significantly lower in tap and surface water, decreasing by approximately 30% and 50%, respectively, compared to ultrapure water⁶⁴.

Conclusions

PFOA was not degraded by the individual action of TiO₂ and H₂O₂ without UV and showed slight degradation with the UV/H₂O₂ and UV/TiO₂ treatments.

PFOA removal with the UV/TiO₂/H₂O₂ system is affected by the H₂O₂ concentration, the TiO₂ concentration, the initial PFOA concentration, and the presence of other substances naturally present in the water.

The addition of H₂O₂ to the UV/TiO₂ system in the present study appears to have a positive effect by slightly accelerating the PFOA degradation process. This behavior is consistent with that observed in other studies in the literature, where the addition of H₂O₂ to the UV/TiO₂ system has been shown to increase the decomposition rate of various organic compounds.

A degradation of approximately 26.08% was achieved after 5 h of treatment in deionized water. The optimal UV/TiO₂/H₂O₂ process conditions tested in surface water showed a degradation of 20.09%.

The 24-hour experiment under optimized conditions showed that PFOA degradation is faster initially, but gradually decreases over time, reaching only 40% removal at the end of the period.

Although in the present study the percentage of PFOA degradation by photocatalysis was not high, the alternative of doping TiO₂ with other photosensitizers to increase the percentage of PFOA degradation could be evaluated. This could be used on a large scale in the treatment of wastewater from industries that use PFOA in the production of cleaning products, firefighting foams, floor polishes, lubricants, etc.

Experiments with untreated surface water showed that the chemical composition of the water could affect the effectiveness of photocatalysis. Organic matter and other components present in this matrix reduce the process efficiency by capturing free radicals or interfering with light adsorption and scattering. Furthermore, organic compounds in higher concentrations than the target contaminant compete for the catalyst’s active sites.

In the future, PFOA degradation tests should be carried out using ozone/UV light 254 nm and ozone/H₂O₂/UV light 254 nm, since they are very effective for the degradation of persistent organic pollutants. Ozone has a high oxidation potential. Peroxone is based on an ozonation process with the addition of hydrogen peroxide. This is done to accelerate the decomposition of the ozone molecule (O₃) into hydroxyl free radicals, which can reduce the reaction time required for the degradation of pollutants.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Gustavo A. Peñuela contributed to the study conception and design. Material preparation, data collection and analysis and experiments were performed by Lizeth Marin. The first draft of the manuscript was written by Lizeth Marin and Gustavo A. Peñuela and Lizeth Marin commented on previous versions of the manuscript. Gustavo A. Peñuela read and approved the final manuscript.

Data availability

The datasets generated or analyzed during this study are available in the institutional repository of the University of Antioquia starting in May 2025. https://bibliotecadigital.udea.edu.co/handle/10495/13.

Declarations

Competing interests

The authors declare no competing interests.