Assessing the Photocatalytic Degradation of Penconazole on TiO₂ in Aqueous Suspensions: Mechanistic and Ecotoxicity Studies in Aerated and Degassed Systems

Abstract

Penconazole (C₁₂H₁₅Cl₂N₃) is widely used to prevent fungal infection of fruits. Since this toxic fungicide is recalcitrant to biological degradation, it has harmful impacts on aquatic ecosystems. TiO₂-based heterogeneous photocatalysis proved to be an efficient method for its mineralization. To monitor the processes occurring under the influence of illumination, the light absorbance, the pH, and the TOC of the samples were measured. The concentration of the model compound and the degradation products were determined by HPLC and IC. Penconazole did not decompose under UV light (λ_max = 371 nm) without a catalyst. In the presence of TiO₂, mineralization took place. The initial degradation rate in air (7.7 × 10⁻⁴ mM s⁻¹) was 5 times higher than under argon. The formation rate of hydrochloric acid (1.04 × 10⁻³ mM s⁻¹) in the former case significantly contributed to the acidification of the liquid phase. NH₄⁺ also formed, at the rate of 5.9 × 10⁻⁴ mM s⁻¹, and very slightly transformed to NO₃⁻. Due to the intermediates identified by HPLC-MS, hydroxylation, H abstraction, and Cl elimination are involved in the degradation mechanism, in which photogenerated HO^● radicals, conduction-band electrons, and (under air) superoxide radical anions (O₂^●−) play considerable roles. The intermediates proved to be much less toxic than penconazole.

1. Introduction

With the increase in agricultural production, more pesticides, fungicides, and herbicides are being used to prevent various plant diseases [1,2]. Their harmful effects on aquatic ecosystems and human health have been widely observed [3,4]. Penconazole (1-[2-(2-(2,4-dichlorophenyl)-pentyl]-1,2,4-triazole) is a fungicide for the prevention of a fungal infection called powdery mildew [5]. It is increasingly used to protect various apple varieties. Apples are the second most popular fruit in the world, so the number of sprays used increases yearly [6]. Due to its poor biodegradation properties, penconazole can persist on plants, enter the surface and groundwater, and accumulate in the environment [7]. It is a carcinogen and is also dangerous for the development of fish eggs, thus damaging aquatic life [8]. Penconazole is a triazole derivative (Figure 1), prepared from 2,4-dichlorophenylbutan-1-one using sodium methylate, chloroacetic acid methyl ester, methane sulfonyl chloride, and triazole [9].

Most of the pesticides, fungicides, and herbicides are recalcitrant to biological degradation. Advanced oxidation processes (AOPs) [10] offer solutions for the mineralization of these pollutants or to transform them into biodegradable derivatives. Some triazole-type fungicides were degraded by the application of magnetite as a catalyst in the presence of added H₂O₂ [11]. A Fenton reaction combined with UV radiation was also applied for the decomposition of such fungicides [12]. However, heterogeneous photocatalysis proved to be one of the most efficient AOPs for this purpose, and TiO₂ is the most widely used catalyst for this purpose [13]. Accordingly, TiO₂-based photocatalysis was successfully applied for the degradation of various fungicides such as tebuconazole [14], carbendazim [15], metalaxyl [16], myclobutanil, penconazole, difenoconazole, and boscalid [17,18], along with other pesticides [19], herbicides, and antibiotics used in agriculture, e.g., atrazine and dimethoate [20] together with chlorpyrifos methyl, heptachlor, and methomyl [21], paraquat [22], and sulfamethoxazole [23]. Nevertheless, silverized TiO₂ did not prove to be more efficient for the degradation of fluconazole than the untreated photocatalyst [24].

Several research groups have studied the degradation of penconazole by photolysis and photocatalysis and the formation of intermediates under different conditions. Hussein and co-workers illuminated solutions containing different concentrations of the fungicide with sunlight and UV light (λ = 254 nm) and observed significant degradation in 4 h, but no intermediate formation was identified [25]. The process has been studied in more detail by Voigt and co-workers, using both UVA and UVC light sources [26,27]. Although the emission maximum of the UVC light source (254 nm) was outside the absorption range of penconazole (200–230 nm), the compound was nevertheless transformed. The authors suggested that degradation in this case may have been indirect, i.e., hydroxyl radicals from the photolysis of water may have triggered the degradation. This finding is supported by the presence of different hydroxypenconazole intermediates in the solution. In the UVA lamp experiments, the fungicide was not degraded, and no hydroxy derivatives were formed. Under UVC illumination, the hydroxylated components accumulated, and no ring opening of the compound was observed in the absence of superoxide radicals. Based on their experimental results, the researchers found that chloride loss is an important step in photodegradation [26]. After the chlorine was removed, the compound underwent a structural transformation, closing the ring. Mineralization can continue only up to a certain point by UV photolysis, as the bonds in the aromatic rings of the compound are difficult to break without the use of a catalyst. This is because ring opening occurs only in aerobic systems or in the presence dditivees (e.g., H₂O₂) that form reactive oxygen-containing particles (e.g., O₂^●−, HO₂^●, O₂(¹Δ_g)) in photocatalytic reactions [28].

Garrido and co-workers studied the sunlight-induced degradation of fungicidal compounds in the presence of ZnO/Fibroin and TiO₂/Fibroin protein particles. The optimal ratio of catalysts to protein components was determined [29]. The current concentration of the contaminant was measured by HPLC analysis. They also used sodium persulfate to increase the degradation efficiency of TiO₂-based photocatalysis [17]. M’Bra and co-workers studied the photocatalytic degradation of penzonazole on both suspended and immobilized TiO₂ [18]. The suspension provided significantly higher efficiency. None of the photocatalytic studies, however, investigated the degradation mechanism under various conditions and how the toxicity of the fungicide changes during the degradation. Only high-energy photolysis was applied for this purpose, such as UVC and vacuum UV irradiation in the case of tebuconazole fungicide [30].

Based on the earlier results, our work focused on studying the TiO₂-based photocatalytic degradation of this fungicide, with special emphasis on mechanistic investigation under various circumstances, e.g., in the absence and presence of dissolved oxygen. For this purpose, complementary analytical methods were applied, e.g., spectrophotometry, IC, HPLC (also combined with MS), TOC and pH measurements. Detection and identification of inorganic and organic compounds produced during decomposition indicated that photogenerated hydroxyl radicals, as well as conduction-band electrons, and, in aerated systems, superoxide radical anions play considerable roles in the mechanism. The intermediates formed under different conditions proved to be significantly less toxic than the starting fungicide itself.

No such comprehensive and systematic study regarding the photocatalytic degradation of penconazole in a suspension of TiO₂ (or other catalysts) without any added oxidizers has been published so far. Therefore, the novelty of our investigations lies in the above-mentioned result, considering that similar works dealt only with the UVC photolysis of this fungicide in aerated systems.

2. Discussion

2.1. Degradation of Penconazole

2.1.1. Photolysis

The photochemical stability of penconazole was determined by UV (LED λ_max = 371 nm) and visible light (LED λ_max = 453 nm). The light absorption spectra of the samples remained unchanged in both cases (see in the Supplementary Materials (abbreviated as SM), Figure S1). The variation in pH and TOC values was within the measurement error. The liquid chromatography analysis data also confirmed these results, i.e., the model compound did not degrade under any of the experimental conditions used in this work, and the variation in concentration was minimal (±3%).

The results agree with those observed by Voigt and co-workers, i.e., the degradability of penconazole is significantly dependent on the wavelength and intensity of the light source [26]. In their experiments, penconazole underwent a structural transformation under UV light (λ_max = 254 nm); hydroxylation by hydroxyl radicals from water. Under longer wavelength (λ_max = 313 nm) and lower energy irradiation, the compound did not decompose. The maximum emission of the UV LED used in our work was at 371 nm, while that of the Vis LED was at 453 nm.

2.1.2. Photocatalysis

Adsorption

In heterogeneous photocatalysis, adsorption is an important step in the degradation of the compound, and therefore the extent of surface binding of the model pollutant was determined in this system. The titanium dioxide catalyst can adsorb organic compounds efficiently due to its high specific surface area (55 m² g⁻¹). The binding of penconazole on the surface of the P25 catalyst was investigated by a 40 min measurement. In the reactor, the reaction mixture was stirred in the absence of light, and air was passed through the solution at a rate of 10 dm³ h⁻¹. After 20 min, the absorbance values decreased by almost 20%, after which the light absorption of the solution changed only slightly (Figure 2). This means that the adsorption–desorption equilibrium is reached in about 20 min.

Comparison of UV and Vis Photocatalysis

After reaching the adsorption–desorption equilibrium, the light absorption of the solution did not change in the aerated system when irradiated with visible light, whereas it decreased rapidly under UV light (Figure 3).

TOC measurement is a suitable analytical method for characterizing photocatalytic processes and allows the decomposition to be well monitored. One of the products of the mineralization of organic compounds is carbon dioxide, which is removed from the solution during the process, i.e., the organic carbon content of the reaction mixture is reduced. Figure 4 compares the change in TOC measured in experiments carried out in the presence of titanium dioxide. When the catalyst was excited with UV light, the organic carbon content of the solution decreased by 70% after 6 h of illumination, whereas with visible light, the change was only 5%, within the measurement error.

The P25 TiO₂ catalyst can only be excited with UV light, and therefore the compound was not modified in the heterogeneous photocatalytic experiments with visible light, even in an aerated system. However, the results clearly demonstrate that the TiO₂ catalyst is an efficient photocatalyst in the UV region, even in the presence of which penconazole can be very efficiently mineralized.

UV Photocatalysis

The photocatalytic degradability of the selected fungicide was investigated under both aerobic and anaerobic conditions. Our aim was to explore the mechanism of degradation in as much detail as possible and to find out what processes can be expected when the photochemical process is carried out in the presence or absence of oxygen.

The actual concentration of penconazole was determined by HPLC analysis. Under an argon atmosphere, the concentration of the model compound decreased by about 46% in 240 min, with an initial decomposition rate of 1.53 × 10⁻⁴ mmol dm⁻³ min⁻¹ (Figure 5). The processes involved were accompanied by a slight acidification of the liquid phase. From an initial value of 4.94, pH decreased to 4.34 in 4 h, i.e., a change of 0.6 (Figure 5).

When the suspension was mixed with air, the concentration of the starting compound decreased rapidly, reaching the limit of detection in 240 min, with an initial degradation rate of 7.68 × 10⁻⁴ mmol dm⁻³ min⁻¹ (Figure 6). This is nearly three times the rate that can be determined under anaerobic conditions. The pH of the liquid phase decreased relatively rapidly (by 1.48) in the first four hours and continued to change a little (Figure 6).

What explains the much lower degradation rate in an argon atmosphere than in the presence of air? One very important reason is that fewer oxidizing radicals are formed. Upon UV irradiation of TiO₂, an electron is excited from the valence band to the conduction band (e⁻_CB), leaving a positively charged hole in the previous one (h⁺_VB). Although the reaction of water with the hole produces hydroxyl radicals (HO^●) (Equation (1)), in the absence of oxygen no other oxygen-containing active radicals (e.g., superoxide radicals, O₂^●−, Equation (2)) are formed that could react effectively with the model compound. Nevertheless, the rate constants of the reactions between nitrogen-containing pesticides and hydroxyl radicals are very high (1.9 × 10⁹–1.2 × 10¹⁰ M⁻¹s⁻¹ [31]). Therefore, it is possible to detect the decomposition process even in the absence of oxygen.

H⁺_VB + H₂O → HO^● + H⁺

(1)

e⁻_CB + O₂ → O₂^●−

(2)

Scheme 1 shows a simplified mechanism for the photocatalytic degradation of penconazole on titanium dioxide.

What causes the acidification of the solution? It is probably the combined effect of several processes. One possible explanation is that the functional groups of penconazole (chlorine atoms attached to the benzene ring and nitrogens in the triazole ring), when released into the solution during degradation, induce the formation of strong acids, hydrochloric and nitric acids. To verify the validity of this hypothesis, the time evolution of chloride and nitrate ion concentrations in the air atmosphere was determined by ion chromatography. Under argon, however, the change in the pH in the solution is small, especially in the first 90 min; thus, chloride ions are hardly formed in this period. Notably, the Increase In the H⁺ concentration during the photocatalysis in aerated suspension was more than one order of magnitude higher than under argon.

In the aerated system, the amount of chloride ions in the solution phase increased rapidly, with an initial formation rate of 1.04 × 10⁻³ mmol dm⁻³ min⁻¹. After 5 h of irradiation, the concentration of the inorganic ion was 0.21 mmol dm⁻³ and then remained unchanged (Figure 7a). This means that the whole Cl content of the starting compound (c(penconazole)₀ = 9.23 × 10⁻² mmol dm⁻³) was transformed to chloride ions, the few percent excesses being due to the error of the IC measurement. At this stage of the experiment, the organic intermediates present in the suspension no longer contain chlorine.

The amount of nitrate ions in the solution increases slowly, with a total concentration of only 0.0064 mmol dm⁻³ after 360 min of illumination and an initial formation rate of 3 × 10⁻⁵ mmol dm⁻³ min⁻¹ (Figure 7b). This result suggests that the degradation of the compound from the nitrogen content does not lead to the formation of nitrate ions in the first step.

In Figure 8, the amount of hydrogen ions calculated from the pH values and the time evolution of the total amount of inorganic ions formed are shown. It can be seen that the trend of concentration changes is almost identical, but it can also be concluded that it is not only the formation of inorganic acids that causes the acidification of the solution phase. The reaction of water molecules adsorbed on the catalyst surface and the photogenerated holes may also contribute to the change in chemistry, producing hydroxyl radicals and hydrogen ions (H⁺) as shown in Equation (1).

Nitrogen Balance

In an air-saturated system, penconazole was virtually completely decomposed in 240 min (Figure 6), while under anaerobic conditions its concentration decreased by about 46% (Figure 5). Only small amounts of nitrate ions were formed from the nitrogen in the triazole ring (Figure 7b). What kind of other nitrogen-containing products can be formed? It is possible that elemental nitrogen is produced and leaves the reaction mixture, or that ammonia is formed in the photocatalytic process and may be present in the solution as ammonium ions. The formation of ammonia was observed in the photocatalytic degradation of other nitrogen-containing pesticides, too, in aerated systems [32]. Ammonia (ammonium ion) may be oxidized to nitrate in photocatalytic reactions.

In the first step, the concentration of the ammonium ion formed was determined by spectrophotometry. The results are presented in Figure 9.

Minimal amounts of NH₄⁺ were formed during degradation under anaerobic conditions. During the photocatalytic experiments in the presence of air, the ammonia concentration initially increased rapidly, with an initial formation rate of 5.92 × 10⁻⁴ mmol dm⁻³ min⁻¹, which is of the same order of magnitude as that of the initial rate of the decrease in penconazole concentration, i.e., 7.68 × 10⁻⁴ mmol dm⁻³ min⁻¹ (see Figure 6). Subsequently, after 300 min of illumination, a very slight decrease in the ammonium concentration was detected in the illuminated sample.

Based on the results obtained so far, it is likely that the formation of ammonia (ammonium ion) and its oxidation to nitrite and nitrate are parallel processes. The probability of ammonia oxidation increases at the fifth hour of illumination when the concentration of the starting compound is already minimal, and the degradation of the intermediates is advanced. This very slow oxidation of NH₄⁺ resulted in a slight decrease in its concentration toward the end of irradiation, along with the monotonous but slow increase in the nitrate concentration (Figure 10).

Figure 11 shows the evolution of the nitrogen balance during photocatalysis in an air-saturated system. The starting solution of penconazole with a concentration of 0.093 mmol dm⁻³ contained three times the amount of nitrogen (0.279 mmol dm⁻³). The nitrogen content of the illuminated samples was calculated considering the actual concentration of penconazole and the determined concentrations of ammonium ion and nitrate ion. The nitrogen balance obtained in this way shows significant deficits, especially in the middle of the irradiation period, reaching 30–35%. One reason for this is that we could not measure the concentration of the nitrogen-containing organic intermediates produced during degradation (the composition of the intermediates is not known, and no appropriate standards are available). Formation and escape of elemental nitrogen (N₂) could also occur, which could cause the final deficit at the end of the photocatalysis. Nevertheless, the error of the spectrophotometric method for the determination of ammonia might partly contribute to this deficit.

Spectral Changes

The spectra in Figure 12a clearly show that the light absorption of samples illuminated under aerobic conditions decreases in the 200–240 nm range. This indicates a significant transformation of the starting compound. It can also be observed that the shape of the light absorption curves was slightly modified, suggesting that the intermediates formed during degradation also show light absorption in this range. After 240 min of irradiation, the absorbance is low, i.e., the amount of light-absorbing compounds in the reaction mixture is small. The absorbance depletion in this range also indicates the cleavage of the aromatic ring, promoting total mineralization. In experiments under an argon atmosphere, the light absorption of the irradiated samples displays a modest change in the range of 200–240 nm on this absorbance scale (Figure 12b); for example, at 220 nm, it decreased by 0.147 within the 240 min photocatalysis. Behind these relatively small spectral changes, however, a considerable (46%) degradation of penconazole took place (see Figure 5). The spectra of the hydroxy derivatives formed during the degradation of this fungicide under argon (see later in the section “Identification of intermediates”) are very similar to that of penconazole, as we observed earlier in the case of hydroxylation of the aromatic moiety of other organic pollutants [33]. Hence, their formation moderately modified the absorption spectrum of the solution phase.

In the light absorption spectra of the compound in the range 250–320 nm, the band corresponding to the transition π→π* characteristic of the aromatic system showed peculiar change upon illumination of the reaction mixture (Figure 13a) in experiments under air. The light absorption increased, the fine band structure gradually disappeared, a new shoulder appeared at 295 nm in addition to the previous local maxima at 272 and 281 nm, and then the light absorption decreased after 120 min. In anaerobic photocatalysis, the spectral changes were smaller in the range investigated (250–320 nm), with light absorption increasing gradually throughout the 240 min exposure (Figure 13b).

For the bands corresponding to the model compound (220 nm and 272 nm) and for the “new shoulder” (295 nm) typical of the intermediates formed during photocatalysis, the changes in the absorbances under different experimental conditions are compared in Figure 14.

The increase in the absorbance in the range typical of the aromatic ring is because the light absorption of the resulting intermediates, presumably “hydroxy-aromatic compounds”, is more intense than that of the starting compound. The increase in their concentration, i.e., in their light absorption, compensates for the decrease in the original band at 272 nm. The increase in absorbance around 295 nm probably indicates the formation of hydroxypenconazole intermediates. All this suggests that hydroxy derivatives are formed more slowly in an argon atmosphere and only accumulate (Figure 14b), whereas they form and decompose more rapidly in an air-saturated system (Figure 14a).

How do we assume that the spectral changes indicate hydroxylation of the aromatic ring? Members of our research group have previously studied in detail the photocatalytic degradation of phenylalanine [28] and benzenesulfonic acid [34]; in each case, similar spectral changes were observed. The identification of the intermediates formed was facilitated by the commercial availability of the hydroxy derivatives (4-hydroxyphenylalanine (tyrosine) and 4-hydroxybenzenesulfonic acid). By recording their light absorption spectra and calculating the molar light absorption coefficients, it was found to be true in both cases that the molar light absorption coefficients of the hydroxylated compounds were significantly higher than those of the parent compound [28,34].

Mineralization and Intermediates

The decrease in the TOC in the photocatalyzed system indicates the advance of the mineralization of the organic components. The spectral changes, the formation of inorganic ions, and the pH variation of the solution suggest the formation and degradation of various organic intermediates during photocatalysis. The TOC value for penconazole was calculated from the actual concentration of the unreacted fungicide. Its subtraction from the total organic carbon content of the solution provided the TOC value represented by the intermediates over time. Figure 15 clearly shows that the mineralization of the starting compound was slower in an argon atmosphere than in air-saturated systems. In the latter case, the TOC content of the starting material practically reached the limit of detection already at the 240th minute. Initial rates of TOC depletion:

In argon atmosphere: 0.0158 mg dm⁻³ min⁻¹;

In air atmosphere: 0.0332 mg dm⁻³ min⁻¹.

During the 4 h illumination, the carbonaceous intermediates are only formed in the argon atmosphere, whereas in the presence of air, they decrease after 120–180 min, i.e., they degrade rapidly.

Identification of Intermediates

The actual concentration of penconazole was determined by HPLC analysis, and in addition to the 9.3 min retention time (t_R) peak (HPLC-1) typical of the starting material, new peaks were observed in the chromatograms of the irradiated samples. The unknown compounds were not available as standards, so no calibration curves could be obtained. Hence, the exact concentrations cannot be calculated from the peak areas, but their magnitudes are proportional to the amount of the corresponding component.

During the photocatalysis in the argon atmosphere, the concentration of penconazole significantly decreased (by about 46%, Figure 5), resulting in the formation of considerable amounts of intermediates (Figure 15). Consistent with these results, only a few intermediates (t_R = 6.46 min, 7.33 min, 8.17 min) appeared in the chromatograms. As MS identifications indicated, all these intermediates were hydroxypenconazole isomers (see in the next paragraph). The amount of the intermediates increased gradually during 240 min of illumination, and their degradation did not occur. Photochemical processes in the presence of air also produced these intermediates (Figure S2), which not only formed but also degraded rapidly.

Samples from photocatalysis in aerated systems were used to identify intermediates. Measurements were performed by the HPLC-2 instrument coupled with a mass spectrometer. The peak retention time of the resulting chromatograms for penconazole was 8.05 min. The m/z value of the compounds appearing in the chromatograms at retention times of 5.46 min (HPLC-1: 6.46 min); 6.43 min (HPLC-1: 7.33 min); 6.79 min (HPLC-1: 8.17 min) was 301.18 (positive mode), giving a molecular weight of 300 g mol⁻¹ when z = 1 (Figure S3). Notably, a fourth intermediate with the same molecular weight was also detected with rt = 5.41 min. The four intermediates (m/z = 301.18) are structural isomers of each other, formed from penconazole (molar mass 284 g mol⁻¹) by hydroxylation. The mass spectra of the intermediates detected by HPLC-MS are shown in Figure S4. It is also found that the intermediates, which are rapidly formed in the presence of air, are no longer detectable in the reaction mixture after 240 min. The retention time depends on the position of the functional groups attached to the compound; the more polar the compound, the longer the retention time. Taking all this into account, it can be concluded that the formation of the more polar hydroxypenconazoles was favored (t_R = 6.43 and 6.79 min) (Figure S3). This finding also applies to the processes occurring during photocatalysis in the presence of argon (Figure S2b,c).

Continuing the analysis of the samples, the presence of other intermediates was detected. Intermediates with a m/z value of 299.18 (molar mass 298 g mol⁻¹, Figure S4i,j) are also structural isomers of each other, but in contrast to the previous one, the more polar compound is formed in smaller amounts (Figure S5). The compounds correspond to the former hydroxylated derivatives containing an unsaturated bond. No such components were found in the argon samples. Intermediates with molar mass 316 g mol⁻¹ were also detected at retention times t_R = 5.99 and 6.01 min (Figure S4g,h). These are isomers of dihydroxy derivatives of penconazole.

Of the intermediates that appear in the HPLC chromatograms, those with the 7.44 min and the 5.74 min retention times and molar masses of 258 and 256 g mol⁻¹, respectively, had relatively large peak areas (Figure S6). Both can be considered as rapidly evolving components. They reached their maximum intensity at 60 min (t_R = 7.44 min) and 120 min (t_R = 5.74 min), and their concentrations decreased during the further periods of illumination. At this stage of the research, their structures have not yet been determined.

The intermediate with m/z = 265.3 and t_R = 5.09 min shown in Figure S7 was also identified by Voigt and co-workers [21], and its structure was proposed by Hensen et al. [35]. It was formed through hydroxylation, Cl elimination and H abstraction from penconazole. Alternatively, instead of hydroxylation and Cl elimination, hydrolysis of the chlorine substituent could take place. The formation of this component is supported by our previous results, as we have demonstrated the presence of chloride ions during the photocatalytic degradation of the model compound in an aerated system. In an argon atmosphere, only hydroxylation was observed, and the presence of a chlorine-losing intermediate could not be reliably detected. This is in accordance with the observation that chloride (or hydrochloric acid) hardly formed in the photocatalysis under argon.

Hydroxylation can start on a carbon atom with sp² hybridization in the benzene ring, but it can also occur by substitution at a carbon atom of sp³ in the aliphatic part. The hydroxylation of the compound and the opening of the rings can occur in several different ways due to complex degradation processes.

Based on the intermediates detected, a simplified scheme suggested for the photocatalytic degradation routes of penconazole is shown in Figure 16.

2.2. Stability, Reusability of the Photocatalyst

A crucial factor in practical implementations of photocatalysis is the stability, i.e., reusability, of the photocatalytic material. Thus, the stability of the titanium dioxide catalyst was evaluated by its cyclic use for photocatalytic degradation of penconazole in aerated systems. After each run, the catalyst was centrifuged and rinsed with Milli-Q water. It was then left to dry overnight. Five consecutive cycles were carried out, with 6 h of irradiation for each of them. As Figure S8 indicates, a negligible decrease in the photocatalytic performance was observed until the end of the fifth cycle. This may be attributed to the loss of the catalyst between the consecutive runs.

2.3. Ecotoxicity Results

Whenever a treatment is used to remove or during 30 min of contact. The relative degradation can be determined from the measured luminescence values (Equation (3) in Section 3.4.6), and the results are compared in Figure 17.

When the penconazole solution was illuminated with UV light in the absence of TiO₂, there was no degradation, and the concentration of the compound did not change (Figure S1), and consequently, the toxicity of the samples did not decrease (Figure 16). In heterogeneous photocatalytic experiments, the concentration of the fungicide varied depending on the gas atmosphere used (Figure 5 and Figure 6), while intermediates were formed. In all cases, the toxicity of the illuminated samples decreased progressively. This can be explained by the fact that, on the one hand, the concentration of the initial toxic fungicide was reduced and, on the other hand, the intermediates produced were much less toxic for Vibrio fischeri bacteria. These results are in accordance with the observation of Voigt et al. regarding the UVC photolysis of penconazole [27]. It can thus be concluded that the photocatalytic degradation of penconazole is an environmentally friendly process.

3. Materials and Methods

3.1. Materials

The following commercially available analytical grade chemicals were used without further purification: penconazole (C₁₂H₁₅Cl₂N₃) from PESTANAL (Budapest, Hungary), ammonium sulfate, sulfuric acid, CH₃CN (gradient purity, ≥99.9%) from VWR International Kft. (Debrecen, Hungary), Aeroxide P25 TiO₂, NaHCO_3, Na₂CO₃, nitro-prusside sodium (Na₂[Fe(CN)₅NO]·2H₂O) from Sigma-Aldrich Kft. (Budapest, Hungary), phenanthroline (C₁₂H₈N₂), formic acid (HCOOH), potassium-oxalate (K₂C₂O₄H₂O), sodium acetate (CH₃COONa), sodium salicylate (C₇H₅NaO₃), potassium citrate (K₃C₆O₇·H₂O), iron (III) chloride (FeCl₃·6H₂O) from Reanal (Budapest, Hungary), potassium hydrogenphthalate from Merck (Darmstadt, Germany), sodium dichloroisocyanurate from Molar Chemicals (Halásztelek, Hungary). Compressed air was provided from a gas bottle, bubbling into the reaction mixtures [36]. Freeze-dried bacteria (for the Lumistox bacteria test) were purchased from Hach Lange GmbH (Düsseldorf, Germany).

3.2. Penconazole Model Solution

Penconazole solution of 30 mg dm⁻³ was prepared from high-purity solid fungicide. The solution of the measured substance was heated to (50–60 °C) and stirred for several hours (4–6 h). Larger volumes of the solution were prepared at a time and stored in the refrigerator until use. Local maxima in the light absorption spectra of the fungicide: 203 nm, 220 nm, and the band structure typical of an aromatic ring in the range 260–285 nm (Figure S9). The molar absorption coefficients were determined from the absorbance values: ε₂₀₃ = 45318 M⁻¹cm⁻¹, ε₂₂₀ = 10279 M⁻¹cm⁻¹, ε₂₇₂ = 476 M⁻¹cm⁻¹, ε₂₈₁ = 460 M⁻¹cm⁻¹.

The shelf life and stability of the starting solution were investigated over several weeks. A portion of the solution was stored in the laboratory under artificial lighting conditions: in the laboratory but in the dark and in the refrigerator. The light absorbance of the samples was checked weekly. It was found that the solution of penconazole at a given concentration was stable; the compound did not decompose on standing either at room temperature or in the refrigerator and was stable under laboratory lighting conditions. Table S1 compares the results for the refrigerated solutions. The differences between the absorbance values are minimal and within the measurement error. During this work, fresh solutions were prepared every three weeks.

3.3. Photochemical Experiments

3.3.1. Reactor

The experiments were carried out in an 80 cm³ volume reactor made of Duran glass (DWK Life Sciences GmbH, Wertheim, Germany), with one stub for sampling and the other for gas injection. The gas (argon/air) flow rate was 10 dm³ h⁻¹. In the solution, the precipitation of the suspended catalyst was prevented by gas bubbling and magnetic stirring. The UV LED was located at a distance of 10 cm from the reactor (Figure 18a), while the distance between the two Vis LEDs and the reactor was 4–4 cm (Figure 18b). The temperature of the reaction mixture did not change significantly during the illuminations, increasing by a total of 3–4 °C in 240 min. The emission spectra of the light sources are shown in Figure S10. The photon flux entering the reactor was determined by ferrioxalate actinometry [37]:

I (Vis LED λ_max = 453 nm): 5.54 × 10⁻⁴ mol photons dm⁻³ min⁻¹;

I (UV LED λ_max = 371 nm): 2.34 × 10⁻⁴ mol photons dm⁻³ min⁻¹.

3.3.2. Experimental Conditions

For the heterogeneous photocatalytic experiments, the catalyst concentration was 1 g dm⁻³. Based on the volume of the reactor (80 cm³), the required amount of catalyst (80 mg) was ultrasonicated in 10 cm³ of water for 15 min to promote the disintegration of the larger aggregates. A volume of 70 cm³ of the 30 mg dm⁻³ penconazole stock solution was added to the 10 cm³ ultrasonicated suspension (initial penconazole concentration of the solutions to be illuminated: 26.25 mg dm⁻³, 9.23 × 10⁻² mM). After mixing, the suspension was stirred for 20 min in the absence of light before illumination was started to achieve homogenization and adsorption–desorption equilibrium. During the irradiation, 5 cm³ samples were taken at fixed intervals depending on the reaction rate. A Millipore Millex-LCR 0.22 μm filter (Merck Millipore, Burlington, MA, USA) was used to separate the solid phase from the liquid.

In photolysis, the degradability of penconazole was investigated with UV and visible light LEDs under the experimental conditions mentioned above without a catalyst. Then, 70 cm³ 30 mg dm⁻³ of starting solution was diluted with 10 cm³ of high-purity water (Milli-Q) to equalize the initial concentrations.

3.4. Analysis of Samples

3.4.1. UV-Vis Light Absorption and pH Measurements

The light absorption of the samples was analyzed by spectrophotometric analysis. The light absorption spectra of the sample solutions were recorded using a SCINCO S-3100 spectrophotometer (Scinco C. Ltd., Seoul, Republic of Korea) in a quartz cuvette of 1 cm path length.

The pH of the reaction mixture was measured with an SP 10T electrode connected to Consort C561 equipment (Consort, Turnhaut, Belgium). The pH meter was calibrated with pH 7 and pH 4 buffer solutions before each measurement.

3.4.2. HPLC Analysis

Liquid chromatography is a technique for the separation and quantitative analysis of compounds, which is used to determine the actual concentration of the model compound and to separate and identify the intermediates formed. The illuminated samples were analyzed by reversed-phase liquid chromatography. Measurements of fungicide and intermediates were performed using a Shimadzu UFLC-20AD liquid chromatograph (Shimadzu, Kyoto, Japan) (HPLC-1) equipped with a photodiode array detector (PDA). The analysis was performed using gradient elution–eluent A: 2% acetonitrile (ACN) + 98% water + 0.1% HCOOH, eluent B: 100% ACN + 0.1% HCOOH–at a flow rate of 0.5 cm³ min⁻¹. The Phenomenex brand (Torrance, CA, USA) Kinetex C18 column used was 100 × 3 mm with a loading particle size of 2.5 µm. The column temperature was 40 °C, 20 µL of sample was injected, and detection was at 205 nm. The calibration curve with different concentrations of penconazole solutions is shown in Figure S11.

The identification of the intermediates was carried out using another HPLC (HPLC-2) instrument coupled to a mass spectrometer. The instrument was a Waters Acquity UPLC (Waters, Milford, MA, USA) equipped with a PDA and a single quadrupole mass detector QDA detector with electrospray ionization. MS ionization was performed in positive mode. The HPLC equipment involved an automatic injector, a quaternary pump, and a thermostable column space. The retention time of penconazole with the instrument (HPLC-1) was 9.3 min, while for the instrument coupled to the mass spectrometer (HPLC-2), it was 8.05 min. The same column was used for the measurements, and the separation method was the same, so the difference is due to differences in instrumentation, mainly determined by the volume outside the column. The elution program:

0 min: 89% A, 11% B	14 min: 89% A, 11% B
10 min: 5% A, 95% B	18 min: 89% A, 11% B
13 min: 5% A, 95% B

3.4.3. Ion Chromatography

The ion chromatography measurements were carried out with a Dionex DX-500 ion chromatograph (Sunnyvale, CA, USA) equipped with an AS50 automatic sample feeder, a thermostable column stage, and a conductivity detector. KOH with a concentration of 10 mM was used as the mobile phase at a volume flow rate of 0.4 mL min⁻¹. A Dionex IonPac AG14 (4 × 50 mm) guard column was used at 30 °C. In total, 25 µL of sample was added, and the analysis took 22 min. Samples were prepared by diluting 2.5-fold with high-purity water. The chloride and nitrate contents were calculated using a five-point calibration curve from the average of the sample peak areas obtained for two parallel injections.

3.4.4. Total Organic Carbon Measurement

The total organic carbon (TOC) of the samples was determined using a Shimadzu TOC-L. Total carbon (TC) and inorganic carbon (TIC) were measured by injecting 2 × 50 μL samples, the difference of which gives the TOC value. Calibration solutions were prepared from a stock solution of potassium hydrogen phthalate for TC measurements, and from a stock solution containing Na₂CO₃ and NaHCO₃ in the same concentration for TIC measurements.

The measurement is based on the conversion of the carbon content of the samples to CO₂ by catalytic oxidation at high temperature (680 °C), which is detected by so-called non-dispersive infrared (NDIR) detection.

3.4.5. Determination of Ammonia

The ammonia content of the illuminated samples was determined by a spectrophotometric method. Ammonia is present in solution as an ionic form, ammonium ion, which reacts with phenol compounds in an alkaline medium in the presence of a catalyst (e.g., nitro-pusside sodium). A bluish-colored indophenol derivative (the yellow color of the reagent makes the sample green) is formed, the light absorption of which can be measured by photometry.

Reagents used:

-

A: sodium salicylate (C₇H₅NaO₃), tripotassium citrate (K₃C₆O₇·H₂O), aqueous solution of nitro-prusside sodium (Na₂[Fe(CN)₅NO] 2H₂O);

-

B: alkaline solution of sodium dichloroisocyanurate (C₃Cl₂N₃NaO₃·2H₂O).

A volume of 0.2–0.2 cm³ of the two reagents (A and B) was added to the 2 cm³ sample. After waiting 30 min, the light absorption spectra were recorded in a 0.5 cm path length cuvette using a SCINCO S-3100 spectrophotometer (Scinco Co., Ltd., Seoul, South Korea). During this time the samples became green in color; the depth of color is proportional to the concentration of ammonium ions. Aqueous solutions of (NH₄)₂SO₄ were used at different concentrations to prepare the analytical curve (Figure S12).

It was investigated whether penconazole affects the determination of ammonia. Therefore, solutions of (NH₄)₂SO₄-penconazole were prepared with a constant total concentration of (NH₄)₂SO₄ and analyzed in the same way as described in the previous section. The slopes of the two curves are identical, and the difference in absolute values is within the measurement error. The results clearly show (Figure S12) that the nitrogen atoms of the fungicidal compound are not sensitive to the reagents used. The spectrophotometric method is suitable for the analysis of illuminated samples. Several accurate calibrations were performed before each measurement.

3.4.6. Toxicity Measurements

The LUMISTOX Vibrio fischeri bacterial strain produced by Hach Lange Ltd. (Berlin, Germany) was used in the tests. The sample preparation for the antibacterial study was made according to [38]. The frozen putty, which was stored in the freezer until use, contained 20 portions (in small bottles) of luminescent bacteria, 250 cm³ of reconstitution solution, and 50 cm³ of 7.5% (w/w) NaCl solution. A 250 cm³ reconstitution solution was divided into 20 parts so that 12.5 cm³ of solution was available to revive each batch of bacteria. Its exact composition is unknown, based on information from the manufacturer: pH 7 solution containing glucose and sodium chloride. The 7.5% (w/w) NaCl solution was diluted to 2% (w/w) to replace seawater.

The frozen bacteria and the reconstitution solution had to be thawed before measurement. To carry this out, the two glass tubes were placed in about 150 cm³ of water for 10–15 min, the beaker changed the water every 5 min. The thawing time varied according to the ambient temperature. For the revival, 2 cm³ of the 12.5 cm³ reconstitution solution was added to the dried bacterial strain and allowed to stand at 15 °C for 10–15 min. Then, the bacterial suspension was washed into the 15 cm³ glass tube with the remaining reconstitution solution and diluted with a further 1 cm³ of 2% (w/w) NaCl solution. The mixture was then incubated at 15 °C for 40 min. The lifetime of the ‘revived’ bacteria at 15 °C was stated by the manufacturer to be 3 h.

After the incubation period, 300 µL of bacterial suspension was added to 200 µL of the sample volume, and the emission intensity proportional to the viability of the bacteria was measured 30 min after contact. Samples were analyzed with a Toxalert 100 luminometer (Merck, Darmstadt, Germany), which gives the intensity in RLU (Relative Light Unit). In the series of measurements, the reference solution contained 200 µL of Milli-Q water and 300 µL of bacterial suspension. The relative decay of the bacteria was determined according to Equation (3):

$${Relative decomposition}_t\left(\%\right)={\displaystyle \frac{I_{reference\left(t\right)}-I_{sample\left(t\right)}}{I_{reference\left(t\right)}}}\times 100$$

(3)

where I_reference(t) is the emission intensity of the reference or blind sample and I_sample(t) is the emission intensity of the actual sample.

4. Conclusions

The degradation of penconazole fungicide was studied by UV and visible light photolysis and mostly heterogeneous photocatalysis. Irradiation of this fungicide at wavelengths λ_max = 271 and 473 nm in aerated aqueous solution did not cause any appreciable degradation, and no detectable amount of intermediate was formed either. On the Aeroxide P25 TiO₂ photocatalyst surface, almost 20% adsorption was reached in 20 min. No decomposition occurred in heterogeneous photocatalytic experiments under visible light. The concentration of penconazole in the solution phase containing only the active substance was reduced to below the detection range after 4 h (in the presence of air). The initial degradation rate was five times higher than that in the argon-saturated system. Accordingly, during the photocatalytic degradation, the acidification (H⁺ concentration increase) in aerated suspension was more than one order of magnitude higher than under argon. The decrease in pH can partly be attributed to the formation of hydrochloric acid. Notably, a stoichiometric amount of chloride was formed in the aerated system, which, in accordance with the TOC results and spectral changes, indicated total mineralization.

Despite the oxidizing conditions, the nitrogen content of the triazole derivative was mostly transformed to NH₃ (or rather NH₄⁺), and even under air only a small amount of NO₃⁻ was formed very slowly. Considering the nitrogen balance during the photocatalysis in the aerated system, besides these inorganic ions, the formation of organic N-containing intermediates and N₂ can occur.

While under argon, accumulation of mostly hydroxypenconazole derivatives as intermediates took place during the heterogeneous photocatalytic degradation, in an aerated system; formation and, subsequently, total mineralization of further intermediates were detected through hydroxylation, H abstraction, Cl elimination, and ring cleavage. The latter process needs superoxide radical anions. These photogenerated oxidants, along with HO^● radicals and conduction-band electrons, play important roles in the degradation mechanism. The intermediates formed during the decomposition proved to be much less ecotoxic than the original fungicide. All these results indicated that TiO₂-based photocatalysis in aerated systems is an appropriate and environmentally friendly method for the degradation and mineralization of penconazole.