Metaldehyde is best known as the main constituent of slug pellets. This organic compound has been found in relatively high levels in both surface and underground water. While many water treatment processes do not work with metaldehyde degradation, a photocatalytic degradation process has been proved to have a significant effect on metaldehyde stability. Nanosized ZnO/laponite composite (NZLC) was used as a photocatalyst in this investigation of metaldehyde degradation. The reactions were carried out in a ultraviolet C (UVC) lamp fitted batch reactor by considering the following parameters: initial metaldehyde concentration, pH of solution, and light intensity. A comparison of degradation efficiency between photolysis, photocatalysis, and adsorptive ability on NZLC indicated that the latter had the highest efficiency. Furthermore, higher metaldehyde degradation was observed as the initial concentration decreased. However, the fastest metaldehyde degradation rate in heterogeneous photocatalysis was obtained when pH values were greater than 7.0. Consequently, the findings suggest that the removal of metaldehyde by adsorption and photocatalytic degradation using NZLC under UV irradiation was a hybrid reaction process (i.e. photolysis, adsorption, and photocatalysis).
Pesticides are being frequently released into the aquatic environment due to their widespread use in agriculture and gardening. Their toxicity, variety, and persistence can harm the health of an ecosystem directly and threaten the health of human beings by polluting the water sources used for drinking water (Environment Australia 2002). Owing to their high bio-recalcitrant and toxic nature, pesticide compounds in water are of concern to public health (Eriksson et al. 2007). With this in mind, there have been several studies (Zeng et al. 2012; Fenner et al. 2013; Zeng & Arnold 2013) on understanding the rates and pathways of abiotic and biotic transformations that may affect the persistence of pesticides in water bodies and the wider environment.
Metaldehyde, 2,4,6,8-tetramethyl-1,3,5,7-tetraoxacyclooctane (Water UK 2013), is a pesticide commonly used as a molluscicide in gardening and agriculture to control snails and slugs. Owing to the mild, wet climate during winter months in the UK over recent years, the problem of crop damage by slugs and snails is growing. As a result, the usage of slug pellets has increased and relatively high levels of metaldehyde are being found in both surface and drinking waters, at levels above the EU and national standards (EA 2009; DWI 2010; Water UK 2013). Because of its physicochemical properties, metaldehyde cannot be removed efficiently by conventional methods such as granular activated carbon which is usually used to remove organic substances. There have been many studies investigating new techniques to remove metaldehyde from aqueous solutions (Autin et al. 2012; Tao & Fletcher 2013; Busquets et al. 2014). For example, the photodegradation of metaldehyde by ultraviolet (UV)/H2O2 and UV/TiO2 in water was demonstrated (Autin et al. 2012). However, despite some advances on the development of new methods for metaldehyde removal from water, there is still a need for a more cost-effective solution.
TiO2 has always been reported as the most effective metal oxide for use as a photocatalyst (Chen & Ray 1999; Ao & Lee 2003; Doll & Frimmel 2005; Gaya & Abdullah 2008). However, other studies have reported that zinc oxide (ZnO) has exhibited higher photodegradation performance, compared to TiO2 especially under visible light (Behnajady et al. 2006). In addition, ZnO has a similar wide band-gap energy (i.e. 3.2 eV) to that of TiO2, which is suitable for long wavelength applications (Behnajady et al. 2006; Doria et al. 2013). Furthermore, TiO2 powder is too expensive for usage within many recovery processes and reuse cannot be applied as a practical technique for large-scale water treatment plants (Chong et al. 2010). Therefore, a more practical application process is also required.
In another recent study, a newly developed ZnO/laponite composite was applied to degrade metaldehyde (Doria et al. 2013) and humic acid (Kim et al. 2013) in water as a valid alternative to TiO2. This paper builds upon that work and aims to demonstrate the feasibility of nanosized ZnO/laponite composite (NZLC) to degrade metaldehyde without the difficulties in the filtration and recovery processes experienced by photocatalyst powders. The specific objectives are: (1) to evaluate the photocatalytic degradation of metaldehyde in aqueous solution using NZLC; and (2) to determine the effects of the initial concentration of metaldehyde, as well as the light intensity and initial pH on photocatalytic degradation of metaldehyde in aqueous solution.
MATERIALS AND METHODS
Nanosized ZnO/laponite composite
All the chemicals used in this work were of analytical grade. The NZLC were prepared with a specific surface area of 100–120 m2/g and average particle size of 2–3 mm in diameter. Particles of the NZLC composites were mostly comprised of ZnO, but laponite and polyvinyl alcohol (PVA) were also formed in trace quantities. The production process for nanosized ZnO involved mixing with laponite, and PVA was subsequently applied in the presence of boric acid (>1.6 M) to make the new composites stable under extremely low pH and high UV irradiation. A more detailed description of the preparation of NZLC is reported by the authors elsewhere (Kim et al. 2013).
Metaldehyde sample preparation
All of the solutions tested throughout the experiments used millipore grade water. This is because a high number of organic compounds exist in deionized water which may react with the •OH radicals (from the photocatalytic reactions) and would cause considerable interference in the analysis. Metaldehyde PESTANAL from Sigma–Aldrich was used to prepare the stock solution, which was made up to 100 mg/L. For each experiment a different quantity of stock metaldehyde solution was added into a 2 L volume batch photocatalytic reactor and different concentrations of samples were prepared. Metaldehyde of 50 ± 0.5 mg was dissolved in 50 mL of methanol to prepare the stock metaldehyde, which could be stored at a temperature of between 1 and 10 °C for up to 1 year. In this study, 1,4-dichlorobenzene-d4 (1,000 μg/mL SPEX) from Fisher was used as an internal standard. The same amount (10 μL of 2,000 mg/L internal standard solution) was added to both the standard and the samples after being collected into the vial to minimize equipment error.
The concentration of metaldehyde was measured using gas chromatography (Clarus 500, Perkin Elmer) with mass spectrometer capability and a selective ion monitoring mode. All metaldehyde sample solutions were taken from the UV photocatalysis reactor and filtered through 0.45 μm Whatman cellulose nitrate membrane filters (47 mm diameter) before passing through a pre-conditioned styrene divinylbenzene polymer solid phase extraction column (SPE PP, 3ML, 200MG – 50 columns, Scientific & Chemical Supplies Ltd, Bilston, West Midlands, UK) within solid phase extraction (Dionex Auto Trace 280 SPE, Thermo Scientific). After metaldehyde extraction in aqueous phase, the extracted solution was transferred to the auto sampler. Then the concentration of metaldehyde was determined using the parameters listed in Table S1 of the Supplementary Material (available online at http://www.iwaponline.com/ws/015/002.pdf).
Batch photocatalytic reactor
Photocatalytic metaldehyde degradation was carried out in a batch adsorption and photocatalysis reactor, with a stainless steel net inside, holding a maximum four ultraviolet C (UVC) lamps and NZLC balls. The NZLC balls were contained in the outside and the cross in the net, while the remaining four holes were prepared for holding the lamps (Figure S1 of the Supplementary Material, available online at http://www.iwaponline.com/ws/015/002.pdf). The UVC lamps provided a main output of 254 nm (intensity = 2.1 mW/cm2). All parts of the reactor were made from stainless steel in order to enhance the UV light power and repress the organic compounds within the samples reacting with some elements on the surface of the reactor. Aeration conditions were controlled by connecting the system to provide uniform mixing of the solution and to reduce the recombination generated from positive holes and electrons (Yamazaki et al. 2001; Gogate & Pandit 2004). Moreover, a temperature sensor and cooling bath were operated at the bottom of the reactor to prevent the lamps from overheating the metaldehyde solution; hence, the temperature was maintained between 19 and 21 °C. All experiments were performed in triplicate using a metaldehyde solution with different initial concentrations and fixed NZLC loading amount during a reaction period of 60 min based on a preliminary study (Kim et al. 2013).
RESULTS AND DISCUSSION
Characterization of the nano-ZnO/laponite composite
The scanning electron microscope (SEM) images (Figure S2 of the Supplementary Material, available online at http://www.iwaponline.com/ws/015/002.pdf) of the newly developed NZLC were observed to examine its structural and surface morphologies. The SEM images showed that all small particles are agglomerated together to form larger clusters, with a non-uniform particle shape. It can also be seen that the average size of particles was in the range of 100–150 nm (Kim et al. 2013). The porosity characterization of NZLC was shown in Figure S3 of the Supplementary Material (online at http://www.iwaponline.com/ws/015/002.pdf) using transmission electron microscopy with a scale bar of 100 nm. In addition, the X-ray fluorescence (XRF) analysis was performed to determine the elemental composition of NZLC. These results are reported elsewhere (Doria et al. 2013). Each NZLC contains 95% ZnO by weight on average, based on these semi-quantitative indications of NZLC composition from XRF results. The rest of the composition was accounted for by laponite, PVA, and other trace raw materials.
Degradation of metaldehyde using nano-ZnO/laponite composite
The photocatalytic degradation efficiency of metaldehyde was evaluated after batch experimentation using the following conditions; only metaldehyde solution (control test), metaldehyde solution with UV lamps (photolysis), metaldehyde solution with NZLC (adsorption), and metaldehyde solution with UV + NZLC (photolysis + photocatalysis + adsorption). The concentration changes (Figure S4 of the Supplementary Material, online at http://www.iwaponline.com/ws/015/002.pdf) in the control test were insignificant [i.e. C/Co = 0.97–1.02], indicating that both adsorption of metaldehyde to the stainless steel reactor and the partitioning of metaldehyde to headspace were found to be negligible.
According to Figure S4, with only UV irradiation and only NZLC adsorption, metaldehyde removed around 15% and 39%, respectively, during the 60 min reaction time. The sorption of metaldehyde to interlayer surfaces of swelling laponite is non-linear whereas the sorption of metaldehyde to nano-ZnO particles is linear. These results might be due to the differences in accessibility of each mineral surface to metaldehyde, as affected by near-surface geometric effects and by the preferential adsorption of water molecules onto hydrophilic mineral surfaces (Joo et al. 2008). However, the sorption of metaldehyde to crosslinked PVA is linear and thermodynamically favourable due to the hydrophobic partitioning of crosslinked PVA. The relatively low amounts of sorption shown by the hydrophobic organic compounds to the three active components (i.e. laponite, nano-ZnO, and crosslinked PVA) in heterogeneously combined NZLC are consistent with other studies (Joo et al. 2013).
Metaldehyde removal using NZLC under UV irradiation showed around 69% degradation efficiency, which is higher than only UV irradiation and only NZLC adsorption. This indicates that photocatalytic degradation using ZnO can be achieved by direct reaction with photons generated by UV irradiation, and/or by indirect reaction with OH radicals generated by the reaction between the ion and ZnO. It is also likely that the porous surface area of NZLC contributed to an increase in the efficiency for metaldehyde degradation. Therefore, NZLC developed by ZnO with laponite was found to be a prominent photocatalyst and absorbent for photodegradation and adsorption of metaldehyde in aqueous solutions. Several previous studies (Doria et al. 2013; Kim et al. 2013; Jeong et al. 2014) reported that supported ZnO on an adsorbent catalyst exhibits a synergism that has marked effects on the reaction kinetics of organic contaminants and their metabolite removal from water.
Effect of initial concentration of metaldehyde
The removal efficiency of metaldehyde was monitored by varying the initial metaldehyde concentration. This set of experiments was carried out at 0.1, 0.5, 1.0, and 2.0 mg/L in which the pH, light intensity, and the NZLC loading amounts were fixed at 7, 8.4 mW/cm2, and 64 g/L, respectively. A value of 95% of the metaldehyde was removed at 30 min for initial concentrations of 0.1 mg/L whereas only 55% of metaldehyde was removed even at 60 min for initial concentrations of 2 mg/L, indicating that the degradation rate increased as the initial concentration of metaldehyde decreased (Figure 1).
Based on this linear regression, the photocatalytic degradation (k) and adsorption (Kads) of metaldehyde to NZLC can be obtained.
As shown in Figure 1 and Table 1, the Kapp values increased from 0.0129 to 0.0529 min−1 as the initial concentration of metaldehyde decreased from 2 to 0.1 mg/L. An explanation for this behaviour is that as initial concentration of metaldehyde increased, more metaldehyde molecules were adsorbed on the NZLC surface which is well crosslinked with PVA. Thus both hydroxyl radical and oxidants were generated and positive holes in the valence band were suppressed. This would result in a lower degradation rate in high concentrations of metaldehyde solution. In addition, the photons created from UV lamps would be interrupted under these conditions by the high concentration of metaldehyde molecules before they reached the NZLC surface. Similar results have been published in previous studies (Behnajady et al. 2006; Kim et al. 2013).
|Run .||Co (mg/L)a .||Light intensity of UVb (mW/cm2) .||pHc .||Kapp (min−1)d .|
|Effect of initial concentration|
|Effect of light intensity|
|Effect of pH|
|Run .||Co (mg/L)a .||Light intensity of UVb (mW/cm2) .||pHc .||Kapp (min−1)d .|
|Effect of initial concentration|
|Effect of light intensity|
|Effect of pH|
aInitial concentration of metaldehyde.
bUVC (254 nm with minimal portion of 185 nm UV light).
cpH was adjusted as a constant with minimal errors (±0.2) throughout the whole experiments.
dApparent or pseudo-first-order degradation rate constant.
eOnly adsorption applied without UV irradiation.
A satisfactory linear correlation coefficient equal to 0.987 was obtained between 1/Kapp and the initial concentration of metaldehyde (Figure 1). Both adsorption of metaldehyde to NZLC (Kads) and photocatalytic degradation of metaldehyde by ZnO (k) were obtained as 2.120 mg/L/min and 0.0318 L/mg, respectively (see Equation (4)). Different values were given for both k and Kads accounting for the different effects these parameters have on metaldehyde degradation. This implies that a greater value of Kads would contribute considerably to the removal of metaldehyde by adsorption over the 60 min interval.
Effect of light intensity
The influence of light intensity on the degradation efficiency of metaldehyde was performed at different light intensities (i.e. 2.1, 4.2, 6.3, and 8.4 mW/cm2). The light intensity was adjusted by changing the number of UVC lamps switched on within the reactor. Figure S5 of the Supplementary Material (online at http://www.iwaponline.com/ws/015/002.pdf) shows that the relationship between the apparent first-order rate is constant with the light intensity for metaldehyde degradation. The Kapp values increased with increasing light intensity from 0.0099 to 0.0197 min−1 from one lamp to four lamps. This finding suggests that more UV irradiation generates more photons which are required for the electron transfer from the valence band to the conduct band and obstructs the recombination of electron-hole pairs of ZnO in the NZLC, resulting in an increased degradation rate. In addition, adsorption of NZLC without UV irradiation was found to be 0.077 min−1 of Kapp (Table 1) and increased consistently as the number of lamps increased. This indicates that adsorbed metaldehyde molecules on the surface of NZLC were degraded by UV irradiation with ZnO photocatalysis and consequently the degradation rate increased.
Effect of pH in aqueous solutions
The values of Kapp increased with increasing pH up to 11. This result can be attributed to the excess of hydroxyl anions, which facilitates photo generation of •OH radicals in alkaline solution. Since hydroxyl anions are adsorbed on the surface of ZnO, more can be generated and form more •OH radical, the degradation rate of neutral metaldehyde molecules increased at higher pHs. However, the degradation rate of metaldehyde decreased when pH increased. This may be due to slight dissolution of ZnO at low pH (Behnajady et al. 2006). Other findings have shown photocatalytic degradation of pesticides (Korake et al. 2012; Doria et al. 2013) and endocrine disruptors (Pardeshi & Patil 2009) using synthesized ZnO to be more effective at higher pH values due to the enhanced formation of ·OH radicals.
Mechanism of NZLC performance
It is widely known that the photocatalysis mechanism entails initial adsorption of pollutants to the surface of photocatalyst and photocatalytic degradation of adsorbed pollutants on the surface of photocatalyst (Fujishima et al. 2000; Andreozzi et al. 2000). As mentioned above, the adsorptive ability of NZLC is more effective than that of photocatalytic degradation in terms of removing the metaldehyde in water. This is because more metaldehyde was adsorbed to both crosslinked PVA and high-surface-area mineral surfaces (i.e. ZnO and laponite). Figure S7 of the Supplementary Material (online at http://www.iwaponline.com/ws/015/002.pdf) shows the diagrammatic representation of NZLC whereby metaldehyde compounds in aqueous solution are adsorbed by the pores on the surface of NZLC and subsequently undergo photocatalytic degradation by ZnO contained in the NZLC. Therefore, NZLC provides both adsorptive capability and photocatalytic degradation to remove metaldehyde. However, some of the adsorbed metaldehyde compounds were not affected by photocatalytic degradation; hence, they were left adsorbed in the pores contributing to the subsequent saturation within the structure. This phenomenon explains why total removal efficiency of metaldehyde by NZLC was not reached completely.
This research work has evaluated the feasibility of nanosized ZnO/laponite composites (NZLC) as a valid alternative to TiO2 to mineralize metaldehyde without limitations in the separation, filtration, and recovery of photocatalyst. The adsorption and photocatalytic degradation of metaldehyde were studied under various experimental conditions. Based on the comparison of degradation efficiency between photolysis and photocatalysis, the adsorption shown by NZLC with UV had higher efficiency than with UV irradiation alone. Furthermore, higher metaldehyde degradation was observed as the initial concentration decreased due to the suppressed generation of hydroxyl radicals along with other strong oxidants and interrupted photons before reaching the surface of NZLC. Increasing UV light intensity enhanced the metaldehyde degradation rate as more light intensity generated more photons which attacked the metaldehyde molecules directly and helped to obstruct recombination of electron-hole pairs. Also, a faster metaldehyde degradation rate in heterogeneous photocatalysis was obtained when pH values were greater than 7.0. Consequently, the removal of metaldehyde by adsorption and photocatalytic degradation using NZLC under UV irradiation was a hybrid reaction process (i.e. photolysis, adsorption, and photocatalysis). This work has shown that photocatalytic NZLC is a prominent cost-effective process able to degrade metaldehyde and potentially other organic compounds such as pharmaceuticals and industrial compounds.
The authors acknowledge the contributions of Ms Ziwei Cao and Dr Qizhi Zhou to aspects of the research described in this paper.