Homogeneous Photo-Fenton Degradation and Mineralization of Model and Simulated Pesticide Wastewaters in Lab- and Pilot-Scale Reactors

Abstract

The homogeneous photocatalytic degradation of model pesticide clopyralid (CLPR) has been investigated under various experimental setups. Lab-scale experiments under UV-A radiation in an acidic environment showed that the degradation rate generally increased when increasing either Fe³⁺ or H₂O₂ concentration up to a point beyond which (i.e., 100 mg L⁻¹ for peroxide or 7 mg L⁻¹ for ferric ions) Fenton reagents had little or even detrimental effect on degradation. Thus, there is an optimum concentration of Fenton reagents for maximizing treatment performance, beyond which degradation rates are not enhanced. Excessive concentrations of peroxide and/or catalyst may (i) introduce unnecessary treatment costs, (ii) reduce performance due to scavenging effects, and (iii) raise environmental concerns associated with the disposal of, e.g., high concentrations of iron in the receiving water courses. Switching from UV-A to visible light led to similar rates of degradation, i.e., 86% and 82.2%, respectively, after 90 min of reaction, highlighting the potential of using renewable energy, i.e., natural sunlight, to drive the process. Treatment for 120 min also led to 90% mineralization and quantitative release of nitrogen originally present in the pesticide; this was also accompanied by complete elimination of eco-toxicity to Vibrio fischeri. Pilot-scale experiments were performed in a fountain-type reactor using a commercial pesticide formulation containing CLPR. Both the degradation and mineralization rates increased with increasing the intensity of the incident UV-A radiation from 1.88 to 4.03 mW cm⁻². Experiments were also conducted with different liquid volumes, i.e., from 3 to 8 L. Illumination of 5 L wastewater resulted in 80% mineralization after 60 min and this only slightly decreased to 73% at 8 L. Overall, the findings underline the promising perspectives of the application of the treatment method in upgrading the quality of water and liquid waste containing pesticides.

1. Introduction

The growing global population that is projected to reach 9.7 and 11.0 billion by 2050 and 2100, respectively, generates enormous amount of waste, owing to the production of food, energy, feed, fiber, etc. [1,2].

Pesticide pollution is an issue that negatively impacts on the environment and also concerns human beings and wildlife, being a widespread problem [3,4,5]. These contaminants are typically chemically stable, resistant to direct decomposition by sunlight, non-biodegradable, and highly toxic [3,6,7,8]. They are also characterized by relatively high solubility in water, and mobility [9]. Although some common water treatment processes are effective barriers for the abatement of many pesticides, some of them are not readily removed by such technologies and thus advanced treatments are demanded to remove them from the environment [3,4,5,7,8,10].

Clopyralid (3,6–dichloro-2-pyridine-carboxylic acid) (CLPR) is an auxin-mimic-type herbicide, from the polar and nonvolatile chemical class of pyridine compounds. It is a selective herbicide widely applied to control broadleaf weeds in certain crops and turf and provides control of some brush species on rangeland and pastures [11,12]. CLPR is highly soluble in water, and its solubility increases with pH [7,13,14].

It is known to be particularly stable against hydrolysis and photolysis [13]. It has been frequently detected in the environment and even in drinking water [3,5,11,13]. It is hazardous to a number of endangered plant species and beneficial insects and toxic to certain mammals [15]. Due to its resistance to biodegradation, CLPR is hardly removed by conventional water and wastewater treatments [5,7]. Many studies indicate that a wide range of pesticides are readily degradable by means of heterogenous or homogenous photocatalysis [15]. The pretreatment of wastes containing a hard-to-degrade anthropogenic pollutant is currently one of the most critical challenges of environmental engineering [7]. When a photocatalytic process is applied as a pretreatment, the main objective is the breakdown of organic contents into more biodegradable ones, as well as the removal of toxicity [4,16].

Because of their high reactivity and nonselective oxidation, advanced oxidation processes (AOPs) have been investigated for the treatment of emerging organic pollutants over the past 20 years. They have been proposed as alternative water treatment methods aiming for the mineralization of organic molecules, including pesticides [3,5,8,17,18,19,20,21]. Among different AOPs, heterogenous and homogenous solar photocatalytic detoxification methods have been proposed for the treatment of wastewater containing pesticides [22].

We have previously reported on the heterogenous photocatalytic degradation and mineralization of CLPR in aqueous solutions, employing a wide range of commercial photocatalysts, providing a thorough kinetic analysis in terms of degradation and mineralization, elucidating potential degradation pathways and determining ecotoxicity [15]. We have also previously proposed an integrated system based on solar photocatalysis and constructed wetlands and proven that it has the potential to effectively detoxify wastewater containing CLPR. Three photocatalytic methods under solar light were investigated: the photo-Fenton and the ferrioxalate reagent as well as the combination of photo-Fenton with TiO₂ P25 [22]. We have also shown that reduced graphene oxide (rGO), a low-cost, nontoxic material, may serve as a reliable alternative in the enhancement of TiO₂ photocatalytic efficiency in water processing applications, achieving disinfection and the degradation and mineralization of CLPR [23]. All previous studies of our research group on the abatement of CLPR are summarized in

Table 1. Previously published clopyralid studies of our research group, at a glance.

Treatment	Photocatalytic Methods	Level of Implementation	Matrix	Main Outcome	Reference
heterogenous photocatalytic experiments with CLPR solutions containing TiO2 P25, TiO2 UV-100, ZnO, TiO2 Kronos 7000, Kronos 7001 and Kronos 7500 under UV-A irradiation	different operational conditions, e.g., type of photocatalyst, catalyst loading, initial pH and hydrogen peroxide (H2O2) concentration	lab	aqueous solutions of clopyralid	TiO2 mediated photocatalysis has the potential to provide a sustainable solution in the detoxification of wastewater containing recalcitrant pesticides, either alone or in combination with other methods	[16]
photocatalytic oxidation of CLPR in a bench-scale photocatalytic reactor, equipped with a central UV-A lamp	degradation and mineralization experiments in the presence of three reduced graphene oxide (rGO)/TiO2 composites containing a 1, 5 or 10 wt% nominal content in rGO and bare P25 for comparison purposes	lab	aqueous solutions of clopyralid	rGO, a low-cost, nontoxic material may serve as a reliable alternative in the enhancement of TiO2 photocatalytic efficiency in water processing applications	[17]
solar photocatalytic oxidation and constructed wetlands	photo-Fenton and ferrioxalate reagent, and combination of photo-Fenton with TiO2 P25	pilot	simulated wastewater	the integrated system effectively detoxified wastewater containing clopyralid	[23]

.

In the present study, we focused on the homogeneous photocatalytic process and studied different factors that affect process efficiency (concentration of H₂O₂ and Fe³⁺ ions, type of irradiation (visible and UV-A)). We also studied the extent of photocatalytic mineralization (DOC and total nitrogen), and the toxicity of CLPR solution. Additionally, we assessed the homogeneous photocatalytic oxidation of simulated wastewater containing CLPR in a pilot scale photoreactor under artificial light and studied the effects of the intensity of the incident UV-A radiation, and of the illuminated volume.

2. Results and Discussion

2.1. Homogeneous Photocatalytic Oxidation of Model CLPR Pesticide in Laboratory Reactor

2.1.1. Effect of H₂O₂ and Fe³⁺ on Degradation Efficiency

The concentration of the reagents in the photo-Fenton method is of vital importance, since it may not only affect the costs associated with the size of the photoreactor, but also the operating costs. The addition of high amounts of hydrogen peroxide can unnecessarily increase operating costs since: (i) reaction rates may be reduced, and (ii) peroxide is usually the single most expensive operating parameter in the photo-Fenton and in similar methods.

Photocatalytic degradation of 40 mg L⁻¹ CLPR with photo-Fenton reagent in the presence of 7 mg L⁻¹ Fe³⁺ and different H₂O₂ concentrations (200 mg L⁻¹, 150 mg L⁻¹, 100 mg L⁻¹, 75 mg L⁻¹, 50 mg L⁻¹, 25 mg L⁻¹, 0 mg L⁻¹) is shown in Figure 1A. The effect of H₂O₂ concentration on the initial degradation rate is shown in Figure 1B. In the case of UV-A radiation, the rate increases as the H₂O₂ concentration increases to about 100 mg L⁻¹. However, increasing the H₂O₂ concentration from 100 to 200 mg L⁻¹ has negative consequences, as at high concentrations H₂O₂ may act as an OH scavenger, as shown below (1):

HO • + H₂O₂ → H₂O + HO •

(1)

In order to study the effect of iron ions on the degradation kinetic, experiments were performed at different initial Fe³⁺ concentrations. As shown in Figure 2A, increasing the Fe³⁺ concentration to 7 mg L⁻¹ leads to an increase in the degradation rate. As trivalent iron ions are photo-reduced into divalent ions, which are the major species catalyzing hydrogen peroxide decomposition, the amount of added ferric ions is proportional to the amount of produced hydroxide radicals and hence to the decomposition rate of CLPR. However, increasing Fe³⁺ concentration from 7 to 28 mg L⁻¹ did not appear to significantly affect the initial oxidation rate (Figure 2B). This is explained by the fact that for Fe³⁺ concentrations higher than 7 mg L⁻¹ the transmission of light is impeded due to the high concentration of iron ions, and consequently it is also impeded in Fe³⁺ photoreduction with oxidation products, thus reducing iron concentration in soluble form, which is capable of causing oxidation.

2.1.2. Effect of Irradiation Type

CLPR photocatalytic oxidation under visible radiation and in the presence of the photo-Fenton reagent was promising, since despite the relatively reduced initial degradation rate compared to that under UV-A, it eventually led to similar rates of CLPR degradation after 90 min of irradiation (Figure 3, 82.2% and 86%, respectively). This underscores the significant advantage of photo-Fenton reagent in decomposing organic pollutants into liquid waste under sunlight. Trivalent iron ions also absorb part of the visible spectrum in parallel with ultraviolet radiation.

2.1.3. Mineralization

Under the previously identified optimal conditions, the conversion of organic carbon to inorganic was monitored (Figure 4). Pesticide mineralization in the presence of photo-Fenton reagent was found to be very fast, reaching 90% within 120 min under irradiation (100 mg L⁻¹ H₂O₂, 7 mg L⁻¹ Fe³⁺, UV-A). The change in total nitrogen concentration (as a sum of NO₃⁻, NH⁴⁺, and NO₂⁻) over time, as a percentage of the theoretical nitrogen contained in CLPR, under various experimental conditions, is shown in Figure 4. In all cases, ammonia and nitrate anions were the predominantly formed species, while the amounts of NO₂⁻ were below the detection limit of the method and after prolonged illumination they were oxidized to NO₃⁻. The nitrogen conversion of the pyridine ring of the CLPR molecule, under the given experimental conditions, occurs rapidly, as 100% of the total nitrogen of the molecule is released within 90 min of illumination.

2.1.4. Determination of Acute Toxicity

To determine the toxicity of the CLPR solution during the homogeneous photocatalytic process, samples were collected at selected time intervals during treatment with photo-Fenton reagent (7 mg L⁻¹ Fe³⁺, 100 mg L⁻¹ H₂O₂, pH: 3.0) and UV-A irradiation and their toxicity was assessed on the basis of the inhibition of bioluminescence of V. fischeri bacteria (Figure 5). The initial CLPR solution (40 mg L⁻¹) was found to be highly toxic to these microorganisms, causing 89% inhibition of bioluminescence after 15 min of incubation. The profile of the photocatalytically treated samples showed that there is a declining trend in their toxicity, up to 90 min of illumination (7%) while up to 180 min it remains constant. Toxicity is then nullified, which underscores the efficacy of the photo-Fenton reagent in completely eliminating CLPR toxicity under these experimental conditions.

2.2. Homogeneous Photocatalytic Oxidation of Simulated Wastewater Containing CLPR in the Pilot-Scale Photoreactor

2.2.1. Effect of Intensity of Incident UV-A Radiation

The study on the degradation and mineralization of the simulated wastewater was also focused on the homogeneous photocatalytic oxidation with the photo-Fenton reagent (Fe³⁺/H₂O₂/UV-A, Vis/pH: 3.0) under artificial (UV-A) non-solar illumination. More specifically, the effect of the intensity of the incident UV-A radiation was studied, in the presence of the photo-Fenton reagent (7 mg L⁻¹ Fe³⁺, 100 mg L⁻¹ H₂O₂, pH: 3.0). The intensity of the radiation in the pilot photocatalytic system can be changed by varying the distance of the light sources (lamps) from the surface of the photocatalytic treatment tank. Experiments were conducted under five different radiation intensities, and it was found that by progressively increasing the intensity, an increase in both the initial degradation rate (Figure 6A) and the reaction rate constant k (Figure 6B) occurs.

Moreover, similar conclusions were drawn from the study of the mineralization of the simulated wastewater carried out through the determination of the dissolved organic carbon (DOC) of samples collected at different time intervals (Figure 7A), since the increase in the incident radiation from 1.88 mW cm⁻² to 4.03 mW cm⁻² led to a fourfold increase in the rate constant of the mineralization, k_DOC (Figure 7B).

2.2.2. Effect of the Illuminated Volume

To assess the effect of the illumination of the simulated wastewater in the homogeneous photocatalytic degradation of CLRP with the photo-Fenton reagent, more experiments were carried out. The volume of illuminated wastewater varies by adding suitable overflow rings that are adapted to the photocatalytic treatment tank and allow it to be filled with 3, 5, or 8 L of wastewater. From Figure 8A, it can be seen that increasing the volume of illuminated wastewater from 3 to 8 L has practically no effect on the degradation of CLPR. Regarding mineralization (Figure 8B), it appears that the illumination of 5 L of the simulated wastewater leads to slightly higher percentages of mineralization after 60 min of illumination (80%) compared to the experiments where the illuminated volume was 3 or 8 L (~73%).

3. Materials and Methods

3.1. Materials

Clopyralid (3,6-dichloro-2-pyridine-carboxylic acid, CAS No. 1702-17-6, Mr: 192 g mol⁻¹, Product No: 36758, PESTANAL^®, analytical standard) was a product of Fluka (Sigma-Aldrich Laborchemicalien GmbH, Sellze, Germany) and was used as received. All other reagent-grade chemicals were purchased from Merck and were used without further purification. Doubly deionized water was used throughout the study [14]. For the preparation of simulated wastewater containing clopyralid, used in the pilot scale experiments, appropriate dilutions of a commercial herbicide, Lontrel 100AS^® (active ingredient: clopyralid, 10% w/v, Dow Agrosciences, Indianapolis, IN, USA), with tap water were used.

3.2. Photocatalytic Experiments

Photocatalytic experiments were performed in laboratory-scale and pilot-scale reactors. Laboratory-scale experiments were performed in a closed-type glass cylinder reactor of 500 mL capacity under magnetic stirring, as previously described [14]. Internally, the glass cylinder reactor had a smaller borosilicate glass cylinder adapted to house the light source. The borosilicate glass allows radiation of wavelengths higher than 300 nm to pass. Two types of lamps were used: ultraviolet radiation, region A (320 < hv < 400), with a maximum at 365 nm (OSRAM DULUX 9W/78, UV-A, Munich, Germany); and visible spectrum radiation (400 < hv < 550), with a maximum at 450 nm (OSRAM DULUX S 8/71, Vis, Munich, Germany). After the addition of the pesticide solution to the reactor, its pH was adjusted to 3.0 by adding H₂SO₄ to ensure the presence of Fe³⁺ ions in a soluble form. Subsequently, a sample was collected, Fe³⁺ and H₂O₂ were consecutively added, the solution was illuminated, and samples were collected at certain time intervals and stored in the dark at 25 °C, before any further analysis. Experiments in a pilot-scale reactor were performed as previously described [22], in a unit constituted by three parts: (a) a fountain-type photocatalytic reactor, (b) a reservoir for storing wastewater, which was placed at the bottom part of the reactor, and (c) an Imhoff-type tank. The reactor is illustrated in Figure 9.

3.3. Analytical Procedures

Changes in the concentration of clopyralid were monitored via its characteristic absorption band at 280 nm using a UV-visible spectrophotometer (UV-1700, PharmaSpec, Shimadzu, Kyoto, Japan). Since a linear dependence between the initial concentration of the pesticide and absorption at 280 nm is observed, photodecomposition was monitored spectrophotometrically at this wavelength. Determination of dissolved organic carbon (DOC) was conducted according to standard methods using a total organic carbon analyzer (Shimadzu VCSH 5000, Kyoto, Japan). Inorganic ions were determined in a Shimadzu system consisting of an LC-10 AD pump, a CDD-6A conductometric detector (0.25 μL flow-cell) and a CTO-10A column oven, as previously described [15]. Chromatographic conditions have also previously been described in detail [15,22].

3.4. Toxicity Analysis

Toxicity analysis was performed using the Microtox Model 500 Analyzer (Azur Environmental, Carlsbad, CA, USA), by assessing the bioluminescence of the marine bacteria V. fischeri. Freeze-dried bacteria, reconstitution solution, diluent (2% w/v NaCl), and adjustment solution (non-toxic 22% w/v NaCl) were obtained from Azur. Change in the toxicity of a CLPR solution of initial concentration of 40 mg L⁻¹ during photocatalytic oxidation in the presence of photo-Fenton reagent (7 mg L⁻¹ Fe³⁺, 100 mg L⁻¹ H₂O₂) and UV-A radiation was studied. The inhibition of bioluminescence, compared with a toxic-free control to provide the percentage of inhibition, was calculated using Microtox calculation software.

4. Conclusions

Concerning the homogeneous photocatalytic oxidation of CLPR pesticide in a laboratory photocatalytic reactor, the following conclusions were drawn:

• There usually is an optimum concentration of Fenton reagents to maximize treatment performance, beyond which degradation rates are not enhanced. This has been demonstrated in this work with lab-scale photocatalytic experiments. Excessive concentrations of peroxide and/or catalyst may (i) introduce unnecessary treatment costs, (ii) reduce performance due to scavenging effects, and (iii) raise environmental concerns associated with the disposal of, e.g., high concentrations of iron in the receiving water courses.
• Interestingly, photocatalytic oxidation under visible light resulted in comparable degradation rates to UV-A radiation; this pinpoints the potential of using renewable energy, i.e., natural sunlight, to drive the process.
• Moreover, the proposed process was capable of converting the organic carbon of CLPR to carbon dioxide, as well as the nitrogen content of the pyridine ring to NO₃⁻ and NH₄⁺. Mineralization was accompanied by detoxification as has been demonstrated in experiments on marine bacteria.
• The process was tested in a pilot-scale, fountain-type reactor treating a commercial formulation of the pesticide under consideration. The results were promising, resulting in considerable degradation and mineralization, whose rates increased with increasing light intensity; an increase of the incident radiation from 1.88 mW cm⁻² to 4.03 mW cm⁻² led to a quadrupling of the reaction rate constant k_DOC.
• The increase in the volume of the illuminated wastewater does not practically affect the initial rate of its degradation during the photocatalytic oxidation with the photo-Fenton reagent. Regarding mineralization, illumination of 5 L of simulated wastewater leads to slightly higher mineralization rates after 60 min of illumination (80%), compared to experiments where the illuminated volume was 3 or 8 L (~73%). Thus, wastewater volumes up to 8 L could be treated without efficiency loss, which shows the potential of the process for large-scale applications.