Adsorption of two fungicides on natural clays of Morocco

Applied Clay Science 123 (2016) 37–46

Contents lists available at ScienceDirect

Applied Clay Science journal homepage: www.elsevier.com/locate/clay

Adsorption of two fungicides on natural clays of Morocco Said Azarkan a,b, Aránzazu Peña b, Khalid Draoui a, C. Ignacio Sainz-Díaz b a b

Département de Chimie, Faculté des Sciences, Université Abdelmalek Essaadi, B.P. 2121 M'hannech II, 93002 Tétouan, Morocco Instituto Andaluz de Ciencias de la Tierra (CSIC-UGR), Avda. de las Palmeras, 4, 18100 Armilla, Granada, Spain

a r t i c l e

i n f o

Article history: Received 31 July 2015 Received in revised form 23 December 2015 Accepted 31 December 2015 Available online xxxx Keywords: Metalaxyl Tricyclazole Fungicides Adsorption isotherms Moroccan clay Stevensite

a b s t r a c t North of Morocco is becoming one of suppliers of tourism sites and agricultural products for Europe. Thus, environmental risks from this region are important for the future of these activities. The presence of pesticides in soils and waters can become a serious environmental problem. Clay minerals can be used for mitigation of this problem. In this work, the adsorption of two fungicides, metalaxyl and tricyclazole, on natural Northern Moroccan clays was investigated using the batch equilibration method and several techniques, such as, X-ray diffraction (XRD), thermal analysis (TGA/DSC), and surface area measurement (BET) were used for clay characterization. The data from kinetic and adsorption studies were fitted to different models. The adsorption kinetics of these fungicides followed a pseudo-second-order model. Adsorption data, higher for metalaxyl than for tricyclazole, were fitted the Freundlich model. A natural stevensite was the clay with the highest adsorption capacities for both fungicides. © 2015 Elsevier B.V. All rights reserved.

1. Introduction The widespread use and disposal of pesticides in agricultural activities can produce environmental contamination of surface and groundwater with a possible health risk, due to their potential toxic, carcinogenic and mutagenic effects (Zhao et al., 2008). Trace levels of pesticide residues can be present in soil, water, air and sometimes food, and may result in harmful effects on human and environmental health (Amweg et al., 2005). Besides, these polluted soil components can be present in aerosols suspended in the air during long periods of time and over long distances. Silicates from the Sahara have been detected in North Europe and America (Iuga et al., 2010). The presence of these silicates with organic pollutants in these aerosols may affect atmospheric and stratospheric chemical reactions and their impact on Climate Change may be enhanced by the high levels of solar radiation (Iuga et al., 2010). The increasing presence of pesticides in natural ecosystems (Sánchez-González et al., 2013) has stimulated research on sorbent materials (Zheng et al., 2010; Chaara et al., 2011) which can be used to remediate and prevent soil and water contamination (Park et al., 2014; Stipicevic et al., 2014; Ouali et al., 2015). Immobilization processes occurring in a soil are of great environmental importance because they may lead to a considerable reduction in the bioavailability of organic pollutants, which may be bound to soil constituents by physical forces or chemical reactions of varying strength. These naturally occurring processes are believed to result in pollutant detoxification and have long been suggested for decontamination (Bollag et al., 2002). However despite the role that soil plays as a barrier to attenuate the wide spread of pollutants, different systems have been sought, which may help in pollutant detoxification/mitigation: biological

http://dx.doi.org/10.1016/j.clay.2015.12.036 0169-1317/© 2015 Elsevier B.V. All rights reserved.

degradation using selected microorganisms (Chelinho et al., 2010; Chen et al., 2011), or chemical pathways such as electrochemical oxidation, photocatalytic degradation, or advanced oxidation processes (Lu et al., 2011; Liu et al., 2012; Wang et al., 2012). Another approach is based on the use of physical barriers to increase pollutant trapping, among which the use of natural clays has been proposed because this natural material is highly compatible with environmental policies (Gonzalez-Pradas et al., 1999; Stringfellow et al., 2011; Sánchez-Jiménez et al., 2012). In order to improve the ability of clays to retain pollutants, modifications have been incorporated to obtain other environmentally compatible products (Rodríguez-Cruz et al., 2008; Suciu et al., 2011). In soils the fractions with the highest adsorption capacity for pollutants are the clay fraction along with organic matter. Clay minerals, especially the smectite family, have received considerable attention owing to their physicochemical properties (Bergaya et al., 2006) and high capacity to retain organic compounds. Phyllosilicates are a very abundant group in soils. They have a layered structure and many physicochemical properties of these minerals depend on cation substitutions in the tetrahedral or octahedral sheets, since they confer a residual negative charge to the sheets, which must be compensated by cations in the interlayer space. Because of their layered structure, high specific surface and natural occurrence, clay minerals play a major role in the adsorption of pollutants in soils, especially organic compounds. Sediments in the North of Morocco have interesting amounts of clay mineral deposits especially in the extreme West of the Riff mountains ridge, in the area of Tanger-Tetuan. The main uses of these deposits are as construction materials. In fact, only the region of Tetuan produces more than the 45% of the total clay materials for construction in Morocco. One Moroccan clay with special properties and highly used

38

S. Azarkan et al. / Applied Clay Science 123 (2016) 37–46

in Cosmetic and Health treatments is the one called locally Rhassoul, whose deposit is located in the east side of the middle-Atlas mountains of Morocco. This clay has been identified as stevensite (Chahi et al., 1999). Moreover, the agricultural use in this area is very intensive, which can increase enormously the risk of concentration of pesticides. This issue affects Public Health because this pollution can reach the water-table of drinking and irrigation water, and go into the human cycle. This point has also an international dimension because the North of Morocco is becoming one of the main suppliers of agricultural products to Europe. Besides, this region is becoming highly attractive for tourism and the environmental risks may affect this economic activity (El Bakouri et al., 2007; Azejjel et al., 2010). Therefore, a deep knowledge of the clay minerals and their interactions with pollutants in this area of North of Morocco is essential for Moroccan and European interests and is necessary in order to prevent Public Health risks of soil contamination and for exploring new strategies for a sustainable development of this area. In this study, metalaxyl and tricyclazole were chosen as model compounds of pollutants in water owing to their generalized use to control fungal diseases in a variety of crops. The adsorption of both fungicides, with different physical–chemical properties, onto several samples of Moroccan clay minerals was assessed. One of the aims of this work is to use natural clays for ecological applications hence the use of surfactants with these minerals was not considered in this work to avoid additional pollution. In addition, relationships between pesticide retention and clay structure, identified by using a wide array of instrumental techniques, have been proposed. 2. Materials and methods 2.1. Natural clays samples Four clays with different characteristics from various locations in Northern Morocco (30 cm-depth) were selected: white Bentonite (BT) from Segangane-Nador (57.4%, 24.7% and 12.3% of clay, silt and sand, respectively), Ghassoul clay (ST) from Missour-Meknes (mainly stevensite), a sample from the Dchiriyine zone in Tetouan (CTe) (Azejjel et al., 2009) and a sample from Targuist (CTa) (31.3%, 29.6% and 35.9% clay, silt and sand, respectively). 2.2. Pesticides Two organic molecules were evaluated for the study of adsorption onto clays: metalaxyl and tricyclazole (Fig. 1). Metalaxyl (methyl N(methoxyacetyl)-N-(2,6-xylyl)-DL-alaninate) is a phenylamide fungicide that inhibits the protein synthesis in fungi, by interference with the synthesis of ribosomal RNA. Tricyclazole (5-methyl-1,2,4triazolo[3,4-b][1,3]benzothiazole) is a systemic fungicide which acts as

Fig. 1. Chemical structure of metalaxyl and tricyclazole.

melanin biosynthesis inhibitor (Tomlin, 2003). Standards of both pesticides (N98% purity, Dr. Ehrenstorfer, Germany) were employed in the adsorption study. Their octanol/water partition coefficients (log Kow) are 1.75 and 1.40, and their solubility in water 8.4 and 1.6 mg L−1 for metalaxyl and tricyclazole, respectively, with greater water solubility for metalaxyl, in accordance with its higher content of oxygenated polar groups (Fig. 1).

2.3. Analytical methods The pH of these raw samples was measured in water/clay dispersions, ratio 1:2.5 (w:v). The organic carbon (OC) content was determined by a modified Walkey & Black method (Mingorance et al., 2007). The cation exchange capacity (CEC) was determined by a method based on the triethylene-tetramine-Cu complex (Meier and Kahr, 1999). The measurement of the specific clay surface was carried out with a “Micrometrics ASAP 2010” analyser that uses the process of multilayer adsorption of nitrogen gas (99.98% purity) at 77 K according to the BET theory (Brunauer et al., 1938), based on determining the amount of gas needed to set a monolayer on the solid surface. All water and gases molecules deposited on the sample surface were evacuated, by degassing overnight at reduced pressure (b10−4 Torr) and 120 °C. The chemical composition of clay samples were characterized by X-ray fluorescence (XRF), using a S4 Pioneer BRUKER analyser, after drying the samples at 60 °C for 48 h, and grinding them into a fine powder. The thermogravimetric analysis (TGA) and Differential Scanning Calorimetry (DSC) studies were performed in a Shimadzu 50 H analyser, using Al2O3 as reference, with a heating rate of 20 °C min−1 in air atmosphere. The transmission electron-microscopy (TEM) was performed with a Phillips CM-20 microscope, operated at 200 kV and equipped with a device for analytical electron microscopy (EDAX). The powder X-ray diffraction (XRD) studies were performed by using a Siemens D-5000 diffractometer with Ni filter and the Kα radiation of Cu (λ = 1.544 Å) in the theta range of 3–70° with steps of 0.017° and 19.7 s of counting time per step and a slit width of 0.25°. These diffractograms were recorded as disordered powder sample from raw material and purified samples. The XRD profiles of purified samples were obtained also as oriented aggregate, after treatment with ethylene glycol and after a thermal treatment of 550 °C. For the purification process, samples were sieved at b 2 mm size and suspended in water. An initial treatment with acetic acid was performed to minimize the amount of carbonates, followed by further washings with deionized water; then the sample was suspended in water for 20 min, dispersed with sodium hexametaphosphate, treated in a ultrasonic bath (5 min) and repeated siphoning of the dispersed material (settling 7 h and 22 min at 22 °C, siphoning the supernatant upper 10 cm from atmosphere interface and the bottom residue separated) and the liquid phase was centrifuged. The remaining solid was considered a b2 μm fraction that after drying constituted approximately 20% of the initial sample. For distinguishing the phyllosilicate composition, this purified sample was treated under ethylene glycol vapours during 24 h: After analysing this sample by XRD, the sample was treated at 550 °C during 2 h and a final XRD analysis was performed. The concentration of fungicides in the supernatants was determined by high performance liquid chromatography with diode array detector (HPLC-DAD) (Agilent Series). A 10-μL sample, after filtration by GHP Acrodisc filters (0.45 μm), was injected into a Zorbax RX C8 column (15 cm × 2.1 mm i.d., 5 μm ∅), protected by a guard column Eclipse XDB-C8 (1.25 cm × 2.1 mm i.d., 5 μm ∅) at a flow rate 0.2 mL min−1, the mobile phase consisting of a 50:50 (v:v) mixture of acetonitrile/ water and wavelength detection of 220 nm for metalaxyl and 200 nm for tricyclazole. Calibration was performed by triplicate injection of standard solutions between 0.25 and 8 mg L− 1 (R2 = 0.999). With these conditions, the retention times for metalaxyl and tricyclazole were 4.3 and 3.2 min, respectively.

S. Azarkan et al. / Applied Clay Science 123 (2016) 37–46

Fig. 2. TEM pictures of particles of the BT sample.

39

Fig. 3. TEM micropictures of particles of the ST sample.

The batch equilibration method was also used for the adsorption isotherms. Clay samples were weighed per duplicate into 30-mL pyrex centrifuge tubes, 1 g for BT, 0.1 g for ST, and 0.5 g for CTe and CTa (these amounts are related with the relative adsorption capacity of the adsorbents), to which the aqueous pesticide solutions (20 mL) at concentrations of 0.5, 1, 2, 3, 4 and 5 mg L− 1 were added. After 4-h end-over-end shaking (20 ± 1 °C), a sufficient time according to the preliminary kinetic studies, the samples were centrifuged at 1500 g for 10 min and an aliquot of the supernatant was analysed for pesticide concentration. A control with the pesticide solutions without clay was also run to discard adsorption on the walls or degradation processes. The fungicide adsorption data were fitted to two empirical equations. The Langmuir model assumes that no interaction occurs among adsorbate molecules and that maximum adsorption corresponds to monolayer coverage. The Langmuir equation is expressed as:

2.4. Adsorption kinetics and adsorption isotherms Pesticide adsorption kinetics was carried out using the batch equilibration system, from 2 to 360 min, for a pesticide concentration of 5 mg L−1 at 20 °C. Different mathematical models were employed for the fitting of the experimental data: linear, pseudo-first and pseudosecond order kinetic equations. The pseudo-first order equation can be expressed as: dX ¼ k1 ðX e −X t Þ dt where k1 is the rate constant of the pseudo-first order model (min−1), Xt is the pesticide absorbed amount (μg g−1) at time t (min) and Xe is the absorbed amount at equilibrium. The pseudo-second order kinetic model can be written as,

X max K L C e 1 þ K LCe

dX ¼ k2 ðX e −X t Þ2 dt

Xe ¼

where k2 (μg− 1 g min−1) is the rate constant of the pseudo-second order model.

where Xmax and Xe are the maximum and equilibrium pesticide sorbed amounts (μg g−1), respectively, Ce is the pesticide concentration in

Table 1 X-ray fluorescence analysis of oxide content (%) for the clay samples. Clays

SiO2

Al2O3

Fe2O3

MnO

MgO

CaO

Na2O

K2O

TiO2

P2O5

SO3

SrO

LOIa

total

BT ST CTe CTa

57.32 48.97 52.93 59.07

24.87 4.43 21.17 13.26

3.90 1.87 8.51 3.12

0.00 0.02 0.06 0.06

2.42 20.08 1.97 3.32

1.10 5.38 1.44 4.59

1.90 0.48 0.58 1.96

0.65 1.15 2.08 2.92

0.27 0.26 1.11 0.33

0.04 0.05 0.28 0.10

0.12 3.29 0.30 0.03

0.02 1.19 0.02 0.03

11.36 7.00 9.41 10.86

103.97 94.17 99.86 99.63

a

Loss on ignition: volatile components, water, organic matter, CO2, etc.

40

S. Azarkan et al. / Applied Clay Science 123 (2016) 37–46

where X is the adsorbed pesticide concentration (μg g−1), Kf is the adsorption coefficient (μg1–1/n mL1/n g−1) which represents the degree

Fig. 4. TEM pictures of particles of CTe.

solution at equilibrium (mg L−1) and KL (L mg−1) is the Langmuir adsorption constant. When a dimensionless separation factor, RL ¼

1 ð1 þ K L C 0 Þ

ranges between 0 and 1, the adsorption process is considered favourable. The Freundlich model considers the existence of both monolayer and multilayer adsorption and assumes an energetically heterogeneous adsorbent surface. It is expressed as X ¼ K f  C e 1=n

Fig. 5. TEM micropicture of particles of CTa.

Fig. 6. Powder XRD profiles of our samples, BT (a), ST (b), CTe (c), and CTa (d).

S. Azarkan et al. / Applied Clay Science 123 (2016) 37–46

or extent of adsorption, while the parameter 1/n is an indicator of the system linearity. The Dubinin–Radushkevich (D–R) model was also tested, X ¼ X max eBε

2

where Xmax represents the D–R monolayer capacity (μg g−1), B is the activity coefficient related with the mean adsorption energy (mol2 J−2)

41

and ε is the polanyi potential, which is calculated from the equilibrium concentration Ce, as   1 ε ¼ RT  ln 1 þ C eq The mean adsorption energy, E (kJ mol−1), is derived from parameter B, calculated as E = (2B)−1/2. The magnitude of E is useful for estimating

Fig. 7. Powder XRD profile of samples as oriented-aggregate. Some zones have been enhanced for detailed observation.

42

S. Azarkan et al. / Applied Clay Science 123 (2016) 37–46

the adsorption reaction mechanism. In the case of E b 8.0 kJ mol−1, physical forces may control the adsorption; for E in the range 8–16 kJ mol−1, ion exchange is the working mechanism, while for E N 16 kJ mol−1 adsorption is dominated by particle diffusion (ElShafei et al., 2009). For the statistical analysis of the data, a cluster analysis was undertaken to classify clay properties in relatively homogeneous groups with respect to adsorption parameters (Kf values) using the Ward method as the amalgamation rule and the squared Euclidean distance as metric (Peña et al., 2015).

The relatively high content in Na+ (BT, CTa), and K+ (ST, CTa, CTe) denotes the presence of these ions as interlayer cations. The high content of Fe and Ti in CTe with respect the rest of samples is remarkable. When analysing some particles of each sample by TEM, BT showed a flaky and flat morphology typical of smectites. Besides some prismatic crystals are observed pointing to the possible presence of plagioclase

3. Results and discussion 3.1. Characterization of clay samples In the powder X-ray diffraction analysis, quartz crystals were detected in all samples with a high intense reflection at 27° (2θ units). Calcite reflection at 29° (2θ units) was detected with small intensity in BT and high intensity in CTe and CTa, being negligible in ST. Typical reflections of phyllosilicates were detected at 6.5°, 8.5°, and 12.5° (2θ units) being more intense in BT sample. For a deeper XRD study, these samples were purified to get the b2 μm fraction obtaining 13.8–21.6% of the initial raw material. The XRD of these samples as disordered powder showed the elimination of carbonates due to the acid treatment during the purification process in all samples, although reflections of possible quartz still remain except in BT (Fig. 2). It is remarkable to analyse the (060) reflection that appears at 1.54 Å (62.2°) and 1.51 Å (61°) in BT and ST, respectively, indicating that the BT is a typical dioctahedral smectite and ST is a trioctahedral phyllosilicate, as expected. However, diffractograms of oriented aggregates show a drastic increase of the intensity of the phyllosilicate (001) reflections at 4–8° (2θ units) in all samples except in CTe. In BT the most intense reflection, d(001) = 17–10 Å, corresponds to a hydrated smectite, the reflection at d(001) = 7 Å can be assigned to a small amount of kaolinite, and no quartz and no clear illite/smectite interstratificated was observed. In ST, also the most intense reflection is of the hydrated phyllosilicate with a wide reflection at 3–10°. In CTe the most intense reflections correspond to kaolinite and chlorite at 7.1 and 7.05 Å and the reflection at 25.4° can be assigned to the small amount of anatase (TiO2) corroborated by the Ti content in the chemical analysis (Table 1). After a treatment with ethylene glycol, some shifts of reflections of phyllosilicates are observed in all samples except in CTe remaining the peak at 12.5° corresponding to kaolinite, d(001) = 7.1 Å in all samples. In BT a big expansion of the interlayer space was observed with a shift of the (001) reflexion at d (001) = 17–12 Å with a maximum at 16.8 Å, being similar in ST, and smaller in CTa (Fig. 3) indicating swelling properties of these samples (Fig. S1). After a thermal treatment at 550 °C of the sample-ethylene glycol complexes (Fig. S2), in BT, ST, and CTa, intercalate was eliminated coming back to a dry smectite with a typical spacing d(001) = 9.7–10.1 Å. In CTe, the kaolinite was dehydrated and collapsed, remaining the reflections at 13.7 Å and 9.97 Å of chlorite and illite, respectively. The chemical clay composition is described in Table 1 obtained from X-ray fluorescence analysis. In general a predominance of Si and Al is observed in our natural clay samples. ST shows a strong magnesiccalcic character (20.08% MgO and 5.38% CaO) with higher load of Mg, while for CTe the Fe content is significant (8.51% Fe2O3). Preponderance of Mg in ST (5 times higher than Al) is a typical indication for stevensite with Mg2+ as octahedral cation, in general ICxSi4(Mg3-XAlX)O10 (OH)2, and this composition is similar to that previously reported (Benhammou et al., 2005) with small differences in Si and Ca content. The amount of Ca in CTa is high, due to the presence of carbonate confirmed by the high volatile content (LOI) and XRD. The amount of S and Sr is also high in ST probably due to the presence of sulphates with Sr and Ca, in accordance with a previous report of gypsum presence in Ghassoul clays (Rhouta et al., 2008). The Si/Al proportion varies between 2.04 (BT) and 9.78 (ST), being especially low in BT.

Fig. 8. TGA profile of clay samples.

S. Azarkan et al. / Applied Clay Science 123 (2016) 37–46

ST surface area reveals a high porosity that may be responsible for the high capacity of this material to fix some cations as well as non ionic compounds according to previous studies (Benhammou et al., 2005). After this characterization work, the sample selection of this work represents a set of clays with very different properties. Some are swelling phyllosilicates, a dioctahedral one (BT) with small surface area and high CEC and a trioctahedral one (ST) with high surface area and low CEC, one non-swelling clay (CTe) comprised of mainly kaolinite, chlorite and illite; and one with medium swelling capacity (CTa) and low phyllosilicate amount along with other silicates, like albite, illite, etc.

Table 2 Main characteristics of the studied clays. Clays

pH (water)

pH (KCl)

OCa (%)

CECb (meq/g)

BET surface area (m2 g−1)

BT ST CTe CTa

8.2 8.3 8.6 10.1

6.7 8.1 7.5 8.1

0.21 0.12 0.42 0.04

0.80 0.36 0.13 0.27

42.27 119.07 31.30 33.68

a b

43

Organic carbon content. Cation exchange capacity.

or quartz (Fig. 4). This agrees with data of XRD analysis whose reflections disappeared after purification. The chemical analysis by EDAX showed a composition similar to that of BT in Table 1 for the smectitelike particles. In ST sample, planar flakes were observed, as previously reported for stevensite (Rhouta et al., 2008), along with some gypsum crystals in accordance with the S and Ca contents in Table 1 (Fig. 5). The chemical composition (EDAX) in several stevensite particles is similar to each other. In CTe no flakes were observed and crystals with sharp hexagonal edges of kaolinite were detected along with some small rectangular non-opaque crystals of possible chlorite. Many small dark grains of possibly Fe oxides were observed according to its high Fe content (Fig. 6). The chemical composition of Si is similar in the different particles but the amount of Al, Mg and Fe is more heterogeneous, the Al decreasing when the Mg and Fe content increases. In CTa, a few flakes were observed by TEM and mainly compacted particles, some with sharp edges of kaolinite, other more irregular of possible plagioclases (Fig. 7). The Al content is significantly lower than in BT and CTe, being very variable in the different particles. The thermogravimetric analysis showed a weight loss of 13.1%, 11.3%, 2.7%, and 4.0% for BT, ST, CTe, and CTa, respectively, at T b 250 °C corresponding to water molecules (Fig. 8). This indicates the great water adsorption capacity of BT and ST and the lower capacity for CTe and CTa, due to the illitic character of these last two samples. However, observing the DSC profiles different behaviour of water desorption is observed (Fig. S3). In BT two endothermic peaks can be detected with the minima at 50 °C corresponding to physisorbed water and one shoulder at 140 °C for the water of the interlayer cation hydration, whereas in ST there is one peak at 65 °C and another very intense peak at 104.6 °C indicating a higher amount of strongly bonded water. All clays present a basic pH in water suspensions, all between 8.1 and 8.6 except CTa, which displays a value of 10.1. For suspensions in KCl solutions, the pH is shifted to lower values (6.7–8.1). This behaviour is typical of clay minerals with cationic exchange between the interlayer cations and the external hydronium cation (H3O+) leaving the aqueous medium with excess of OH− anions. This does not happen in KCl solutions. The high pH with ST and CTa may be owing to the presence of carbonates. The cation exchange capacity of the raw samples, without purification, oscillates between 0.80–0.13 meq/g (Table 2). The high value of BT is due to the high concentration of smectite corroborated by the XRD analysis with ethylene glycol, while the low value of CTe is due to the low content in swelling phyllosilicate confirmed above by XRD studies. The clay surface area values range from 31.30 m2 g−1 for the CTe sample to 119.07 m2 g−1 for the stevensite sample. The high

3.2. Adsorption kinetics and adsorption isotherms 3.2.1. Adsorption kinetics Adsorption kinetics for metalaxyl was initially rapid, reaching equilibrium in the first two hours of contact, followed by a slower adsorption, which corresponds to less accessible sites (Table 3 and Fig. 9). The highest value of maximum amount of adsorbed pesticide (Xmax) was found in ST, whereas the samples CTe and CTa show 20 times lower values. These last two samples share similar results, though the rate constant of CTe is significantly higher than in CTa. The Xmax value is surprisingly low in BT taking into account its high CEC. This fact suggests that the adsorption occurs mainly on the external surface and not in the interlayer space. The high value of ST is justified by its high surface area (Table 2). The results from the kinetic study show that the fitting to a pseudo-first order kinetic equation was not possible for any clay. Nevertheless, the data were fitted to a pseudo-second order kinetic model with R2 values ≥0.95, except for CTe (Table 3). Differences between experimental and calculated Xmax values are low for the samples with higher adsorption capacity, ST and CTa (0.27–0.86%), whereas these differences are higher for BT (18.75%) and CTe (28.59%). The value of h0, corresponding to the initial adsorption rate, indicates that initial adsorption was very rapid for ST followed at a great distance by CTe N CTa N BT. In general it can be deduced that fitting to the pseudo-second-order equation was better than to the pseudo-first order model, with higher R2 values and lower differences with experimental Xmax values. Preliminary kinetic studies were performed with tricyclazole obtaining similar results assessing that the contact time of 4 h was sufficient to get the equilibrium and this time can be used for further adsorption isotherms. However, quantitative kinetic calculations could not be determined with tricyclazole due to its lower solubility and low correlation level of the collected data. 3.2.2. Adsorption isotherms The relationship between the pesticide adsorbed and the equilibrium pesticide concentration in the supernatant at 20 °C for all clay samples and both fungicides, metalaxyl and tricyclazole, is described in Fig. 10. The Langmuir equation did not fit the experimental data of metalaxyl adsorption in three out of four clays, and only in two clays for tricyclazole (Table 4). Additionally, the separation factor, RL, points to a favourable adsorption in the fitted cases. On the contrary, adsorption isotherms are all well described by the Freundlich equation with determination coefficients N0.97 for

Table 3 Parameters from the fitting of metalaxyl sorption onto the different clays (±standard error) to the pseudo-first and to the pseudo-second kinetic equation, using the batch method at a pesticide concentration of 5 μg mL−1. METALAXYL Clays

Xmax-exp. (μg g−1)

Pseudo-first order equation Xmax-calc (μg g

BT ST CTe CTa

7.36 ± 0.58 931.4 ± 0.9 28.40 ± 2.01 44.03 ± 0.06

Not fitted 948.5 ± 1.52 Not fitted 46.26 ± 0.56

−1

)

k (min

Pseudo-second order equation −1

)

2

R

0.0089 ± 0.0008

0.761

0.0130 ± 0.0009

0.853

Xmax-calc (μg g−1)

k (g μg−1 min−1)

h0 (μg g−1 min−1)

R2

8.74 ± 0.23 923.4 ± 1.3 36.52 ± 0.60 44.15 ± 0.11

0.061 ± 0.006 0.491 ± 0.045 0.197 ± 0.059 0.023 ± 0.012

4.7 4.2 105 262.7 44.8

0.974 0.959 0.841 0.999

44

S. Azarkan et al. / Applied Clay Science 123 (2016) 37–46

Fig. 9. Kinetic profiles of metalaxyl sorption. Experimental data (symbols) are fitted to the pseudo-second order kinetic equation (lines).

metalaxyl and N0.96 for tricyclazole (Table 4). Adsorption of metalaxyl on the different clays, except for CTe, was higher than that of tricyclazole, in accordance with the physicochemical properties of both pesticides. With respect to the 1/n values, the ones calculated for metalaxyl are close to unity for BT, CTe and CTa (Table 3), i.e., a C isotherm in the Giles classification (Giles et al., 1960). For this isotherm type, the pesticide is partitioned onto the substrate independently of pesticide concentration. For ST, the isotherm is an S type, with 1/n N 1, implying that adsorption increases with adsorbate concentration. This type of isotherm suggests cooperative adsorption in which weak

pesticide-clay interactions are presumed to occur at low aqueous concentrations, whereas at higher concentrations the adsorbed pesticide promotes further adsorption due possibly to the creation of additional adsorption sites through a better alignment of clay platelets and/or condensation of organic compounds in clay interlayers. On the contrary for tricyclazole, the isotherm is of L type (1/n b 1) for CTe and ST. This isotherm indicates a higher clay affinity for the pesticide at low than at high concentrations, indicating a site-limiting adsorption process. Concerning BT the 1/n value is close to unity, suggesting a partition mechanism, while that for CTa corresponds to an S type. This is

Fig. 10. Sorption isotherms of metalaxyl and tricyclazole on the different Moroccan clays. Experimental data (symbols) are fitted to the Freundlich (straight lines) and to the Langmuir models (dotted lines).

S. Azarkan et al. / Applied Clay Science 123 (2016) 37–46

45

Table 4 Parameters of the fitting of experimental data corresponding to the sorption of metalaxyl and tricyclazole on different clays to the Langmuir and Freundlich equations (±standard error). Batch method (0.5–5 mg L−1) at 20 °C and 30 °C. Langmuir

Freundlich

20 °C

20 °C −1

Clays

KL (L mg

BT ST CTe CTa

METALAXYL Not fitted Not fitted 0.10 ± 0.03 Not fitted

BT ST CTe CTa

TRICYCLAZOLE Not fitted 0.23 ± 0.09 0.73 ± 0.16 Not fitted

)

Xmax (μg g

−1

98.2 ± 21.0

2001 ± 451 83.7 ± 7.3

)

RL

0.95

0.90 0.73

2

R

0.987

0.983 0.955

30 °C 1/n

R

Kf (μg1-n mLn g−1)

1/n

R2

5.0 ± 0.3 4566 ± 221 7.9 ± 0.2 7.4 ± 0.8

0.93 ± 0.05 1.66 ± 0.04 0.93 ± 0.02 1.16 ± 0.09

0.984 0.997 0.997 0.975

0.24 ± 0.03 4753 ± 266

2.00 ± 0.10 2.11 ± 0.06

0.993 0.996

5.2 ± 0.4

1.38 ± 0.06

0.991

3.4 ± 0.5 365 ± 6 33.4 ± 0.9 5.2 ± 0.4

1.23 ± 0.12 0.84 ± 0.03 0.51 ± 0.03 1.38 ± 0.06

0.958 0.992 0.982 0.991

2.0 ± 0.3 277 ± 3

0.83 ± 0.02 0.83 ± 0.02

0.972 0.998

Kf (μg

1-n

n

mL g

consistent with previous studies of adsorption of tricyclazole on purified (0.2–2.0 μm) Ca-montmorillonite, where adsorption followed a similar S type adsorption mechanism (Weber, 1982). Observing the Fig. 10, the lines corresponding to the Freundlich equation fit quite well the experimental points. It is remarkable the drastically high pesticide adsorption of ST, being higher for metalaxyl than tricyclazole. In BT and CTa the adsorption capacity is much lower than in ST, and similar for both pesticides. In CTe the adsorption is higher than in BT and CTa but still much lower than in ST. It must be noted that in CTe the adsorption capacity is higher for tricyclazole than for metalaxyl, being this pattern opposite to that in ST. The explanation of this different behaviour can be due to the different chemical structure and functional groups of these molecules. The carbonyl groups of metalaxyl can have electrostatic interactions and hydrogen bonds with the OH dangling groups of the external surface of ST, whereas tricyclazole lacks these polar functional groups. On the contrary, tricyclazole has a strong aromatic character and a planar geometry and the interactions with clay surface occur likely more through the π electrons of the aromatic rings and the lone electron pairs of the S atom. In agreement with the kinetic experiments (Xmax, Table 4) BT is the clay with less adsorption capacity for both pesticides, among the substrates employed. A cluster analysis, based on clay properties and Kf values, reveals that the surface area and MgO content were the clay properties more closely related with the adsorption of both pesticides (Fig. S4). Finally, according to the Dubinin–Radushkevich model, the adsorption process of both pesticides occurred through weak physical interactions, since the E values ranged between 1 and 3 kJ mol−1. Physical adsorption was also reported as the retention mechanism of tricyclazole on Ca-montmorillonite (Weber, 1982). An additional set of isotherms was carried out with some of the clays at a higher temperature, 30 °C. The results show that for both fungicides the Kf values decreased with increasing temperature. The 1/n values increased for metalaxyl, while for tricyclazole either decreased or remained unchanged. Although an exhaustive study of adsorption isotherms in a wide temperature range is out of the scope of this work, we can extract interesting qualitative thermodynamic properties. The general slight decrease of adsorption in most samples with increasing temperature indicates that the adsorption process is thermodynamically stable following a spontaneous process. A thermal analysis by DSC of the clay samples treated with metalaxyl shows the presence of several small endothermic peaks at the range 200–300 °C that appear in all samples and did not appear in the clays without contact with fungicide. This can be assigned to desorption of organic molecules from mineral surfaces (Fig. S5), nevertheless the amount of fungicide is small and no significant change in the TGA profile is observed before and after the adsorption of metalaxyl (Fig. S6).

−1

)

2

4. Conclusions In this work, two different fungicides have been studied. Both are non-ionic molecules, though their chemical structure is completely different. The metalaxyl molecule has a closely globular structure whereas tricyclazole has a planar geometry. Metalaxyl has polar carbonyl groups with strong acceptors of hydrogen bonds, whereas tricyclazole has no polar functional groups. Metalaxyl has a partial aromatic group, whereas tricyclazole is completely aromatic with a high electron delocalization through the benzene ring, the heterocyclic ring and the electron lone pairs of the bridging S atom (Fig. 1). Therefore, the moleculesurface interactions are different for both fungicides. Although all clay samples studied in this work come from the North of Morocco, clay minerals with different properties have been compared. The content of phyllosilicate in BT, ST, and CTa is high with great swelling properties, whereas this content and swelling are very low in CTe. BT, and CTa have mainly dioctahedral phyllosilicates, while ST is clearly a trioctahedral one. However, the CEC of CTa is much lower than that of BT. In spite of these differences between BT and CTa, the adsorption properties are very similar for both fungicides. However, the pesticide adsorption capacity of these two samples is very low in our conditions in spite of their high swelling property. This means that the adsorption is produced out of the interlayer space in all these clay samples. The interlayers of these phyllosilicates are completely full of water and these non-ionic molecules cannot enter in the interlayer space. The adsorption of metalaxyl on CTe is similar to that on CTa and BT, however, the adsorption of tricyclazole is significantly higher on CTe than on CTa and BT. Probably the high content of Fe in CTe can enhance the π-π interactions with the aromatic electron π cloud and the electron lone pairs of S atom of tricyclazole. The high adsorption capacity of ST for both fungicides is directly related with its high surface area, corroborating the interactions in the external surface of mineral. The high surface area is related with small particle size with a high dangling OH bonds in the edges of crystalline domains facilitating hydrogen bond interaction with the carbonyl groups of metalaxyl rather than the non-polar structure of tricyclazole. In summary, the Ghassoul clay, ST sample, can be useful for remediating soils extracting pesticides and for filtering wastewaters polluted with pesticides.

Acknowledgements This work was financially supported by the “Agencia Española de Cooperación Internacional al Desarrollo (AECID)” as a part of the project A/010590/07 and by the regional Andalusian projects (RNM363 and RNM1897). S. Azarkan thanks for a pre-doctoral grant from AECID.

46

S. Azarkan et al. / Applied Clay Science 123 (2016) 37–46

Authors are thankful to Drs. R. Denoyel, B. Bauluz, and F.J. Huertas for their fruitful discussions. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clay.2015.12.036. References Amweg, E.L., Weston, D.P., Ureda, N.M., 2005. Use and toxicity of pyrethroid pesticides in the Central Valley, California, USA. Environ. Toxicol. Chem. 24, 966–972. Azejjel, H., del Hoyo, C., Draoui, K., Rodríguez-Cruz, M.S., Sánchez-Martín, M.J., 2009. Natural and modified clays from Morocco as sorbents of ionizable herbicides in aqueous medium. Desalination 249, 1151–1158. Azejjel, H., Ordax, J.M., Draoui, K., Rodríguez-Cruz, M.S., Sánchez-Martín, M.J., 2010. Effect of cosolvents on the adsorption of ethofumesate by modified Moroccan bentonite and common clay. Appl. Clay Sci. 49, 120–126. Benhammou, A., Yaacoubi, A., Nibou, L., Tanouti, B., 2005. Adsorption of metal ions onto Moroccan stevensite: kinetic and isotherm studies. J. Colloid Interface Sci. 282, 320–326. Bergaya, F., Lagaly, G., Beneke, K., Lagaly, G., 2006. Handbook of clay science. In: Bergaya, F., Theng, B.K.G. (Eds.), Developments in Clay Science vol. 1. Elsevier, Amsterdam. Bollag, J.M., Strynar, M., Ahn, M.Y., Dec, J., 2002. Characterization of enzymatic or abiotic immobilization of xenobiotics in soil. In: Violante, A., Huang, P.M., Bollag, J.M., Gianfreda, L. (Eds.), Soil Mineral-Organic Matter-Microorganism Interactions and Ecosystem Health 28B. Elsevier, Amsterdam, pp. 289–299. Brunauer, S., Emmett, P.H., Teller, E., 1938. Adsorption of gases in multimolecular layers. J. Am. Chem. Soc. 60, 309–319. Chaara, D., Bruna, F., Ulibarri, M.A., Draoui, K., Barriga, C., Pavlovic, I., 2011. Organo/layered double hydroxide nanohybrids used to remove non ionic pesticides. J. Hazard. Mater. 196, 350–359. Chahi, A., Duringer, P., Ais, M., Bouabdelli, M., Gauthier-Lafaye, F., Fritz, B., 1999. Diagenetic transformation of dolomite into stevensite in lacustrine sediments from Jbel Rhassoul, Morocco. J. Sediment. Res. 69, 1123–1135. Chelinho, S., Moreira-Santos, M., Lima, D., Silva, C., Viana, P., Andre, S., Lopes, I., Ribeiro, R., Fialho, A.M., Viegas, C.A., Sousa, J.P., 2010. Cleanup of atrazine-contaminated soils: ecotoxicological study on the efficacy of a bioremediation tool with Pseudomonas sp ADP. J. Soils Sediments 10, 568–578. Chen, S., Lai, K., Li, Y., Hu, M., Zhang, Y., Zeng, Y., 2011. Biodegradation of deltamethrin and its hydrolysis product 3-phenoxybenzaldehyde by a newly isolated Streptomyces aureus strain HP-S-01. Appl. Microbiol. Biotechnol. 90, 1471–1483. El Bakouri, H., Ouassini, A., Morillo-Aguado, J., Usero-Garcia, J., 2007. Endosulfan sulfate mobility in soil columns and pesticide pollution of groundwater in Northwest Morocco. Water Environ. Res. 79, 2578–2584. ElShafei, G.S., Nasr, I.N., Hassan, A.S.M., Mohammad, S.G.M., 2009. Kinetics and thermodynamics of adsorption of cadufos on soils. J. Hazard. Mater. 172, 1608–1616. Giles, C.H., McEwan, T.H., Nakhwa, S.N., Smith, D., 1960. Studies in adsorption. Part XI. A system of classification of solution adsorption isotherms, and its use in diagnosis of adsorption mechanisms and in measurement of specific surface areas of solids. J. Chem. Soc. 3, 3973–3993. Gonzalez-Pradas, E., Villafranca-Sanchez, M., Socías-Viciana, M., Fernández-Pérez, M., Ureña-Amate, M.D., 1999. Preliminary studies in removing atrazine, isoproturon and imidacloprid from water by natural sepiolite. J. Chem. Technol. Biotechnol. 74, 417–422.

Iuga, C., Sainz-Díaz, C.I., Vivier-Bunge, A., 2010. On the initiated oxidation of C2–C5 aliphatic aldehydes in the presence of mineral aerosols. Geochim. Cosmochim. Acta 74, 3587–3597. Liu, C.S., Shih, K., Sun, C.X., Wang, F., 2012. Oxidative degradation of propachlor by ferrous and copper ion activated persulfate. Sci. Total Environ. 416, 507–512. Lu, L.-A., Ma, Y.-S., Kumar, M., Lin, J.G., 2011. Photochemical degradation of carbofuran and elucidation of removal mechanism. Chem. Eng. J. 166, 150–156. Meier, L.P., Kahr, G., 1999. Determination of the cation exchange capacity (CEC) of clay minerals using the complexes of copper (II) ion with triethylenetetramine and tetraethylenepentamine. Clay Clay Miner. 47, 386–388. Mingorance, M.D., Barahona, E., Fernández-Gálvez, J., 2007. Guidelines for improving organic carbon recovery by the wet oxidation method. Chemosphere 68, 409–413. Ouali, A., Belaroui, L.S., Bengueddach, A., López-Galindo, A., Peña, A., 2015. Fe2O3palygorskite nanoparticles, efficient adsorbates for pesticide removal. Appl. Clay Sci. 115, 67–75. Park, Y., Sun, Z., Ayoko, G.A., Frost, R.L., 2014. Removal of herbicides from aqueous solutions by modified forms of montmorillonite. J. Colloid Interface Sci. 415, 127–132. Peña, A., Mingorance, M.D., Guzmán-Carrizosa, I., Fernández-Espinosa, A.J., 2015. Improving the mine soil quality for a vegetation cover after addition of sewage sludges: Inorganic ions and low-molecular-weight organic acids in the soil solution. J. Environ. Manag. 150, 216–225. Rhouta, B., Kaddami, H., Elbarqy, J., Amjoud, M., Daoudi, L., Maury, F., Senocq, F., Maazouz, A., Gerard, J.-F., 2008. Elucidating the crystal-chemistry of Jbel Rhassoul stevensite (Morocco) by advanced analytical techniques. Clay Miner. 43, 393–403. Rodríguez-Cruz, M.S., Andrades, M.S., Sánchez-Martín, M.J., 2008. Significance of the longchain organic cation structure in the sorption of the penconazole and metalaxyl fungicides by organo clays. J. Hazard. Mater. 160, 200–207. Sánchez-González, S., Pose-Juan, E., Herrero-Hernández, E., Álvarez-Martín, A., SánchezMartín, M.J., Rodríguez-Cruz, S., 2013. Pesticide residues in groundwaters and soils of agricultural areas in the Agueda River Basin from Spain and Portugal. Int. J. Environ. Anal. Chem. 93, 1585–1601. Sánchez-Jiménez, N., Sevilla, M.T., Cuevas, J., Rodríguez, M., Procopio, J.R., 2012. Interaction of organic contaminants with natural clay type geosorbents: potential use as geologic barrier in urban landfill. J. Environ. Manag. 95, S182–S187. Stipicevic, S., Sekovanic, L., Drevenkar, V., 2014. Ability of natural, acid-activated, and surfactant-modified Terra Rossa soils to sorb triazine herbicides and their degradation products. Appl. Clay Sci. 88-89, 56–62. Stringfellow, A.M., Simoes, A., Smallman, D., Beaven, R., Powrie, W., Potter, H.A.B., 2011. Sorption of mecoprop by two clay landfill liner materials: Oxford clay and Mercia mudstone. Q. J. Eng. Geol. Hydrogeol. 44, 321–329. Suciu, N.A., Ferrari, T., Ferrari, F., Trevisan, M., Capri, E., 2011. Pesticide removal from waste spray-tank water by organoclay adsorption after field application to vineyards. Environ. Sci. Pollut. Res. 18, 1374–1383. Tomlin, C.D.S., 2003. The Pesticide Manual. 13th ed. British Crop Protection Council, UK. Wang, Y.K., Tao, L., Chen, M.-J., Li, F.-B., 2012. Effects of the Fe-II/Cu-II interaction on copper aging enhancement and pentachlorophenol reductive transformation in paddy soil. J. Agric. Food Chem. 60, 630–638. Weber, J.B., 1982. Ionization and adsorption–desorption of tricyclazole by soil organic matter, montmorillonite clay, and Cape Fear sandy loam soil. J. Agric. Food Chem. 30, 584–588. Zhao, M.R., Zhang, Y., Liu, W.P., Xu, C., Wang, L.M., Gan, J.Y., 2008. Estrogenic activity of lambda-cyhalothrin in the MCF-7 human breast carcinoma cell line. Environ. Toxicol. Chem. 27, 1194–1200. Zheng, W., Guo, M., Chow, T., Bennett, D.N., Rajagopalan, N., 2010. Sorption properties of greenwaste biochar for two triazine pesticides. J. Hazard. Mater. 181, 121–126.