A clean environment is required for maintaining good health and agricultural sustainability. Regular use of pesticides in modern day agriculture demands the need to devise a means of removing or reducing possible pesticide residues from our environment. Waters that are available in the environment where pesticides are used are of high risk of harboring pesticide residues. Leaching enhances environmental pollution as chemicals drain from the treated region to non-tar- geted environments. Therefore, surface waters have the potential of getting contaminated when irrigation water that has passed over pesticide-treated environment leaches into the surface waters  . Storms could sometime result in spontaneous flow of contaminated water into surface water  . Another source of pollution is drift that occurs if a pesticide spray misses its targets having been deflected by wind or resulting from human error, thereby, landing on a non- targeted area. When the level of the pesticide contamination reaches a critical level in food, ground waters, lakes, rivers or ponds, it becomes an issue that could lead to illness or death in the organisms that depend on them.
Due to its unique attributes, zeolite is a mineral with the potential of removing chemical contaminants from water as earlier published   . Some of the past efforts made in removing pesticide residues in water include the use of clay   , activated carbon   and ozonization  . Use of clay is limited by its adsorption capacity due to its shrink-swell behavior and zeolites are free of such flaws  . Saturation of carbon filters resulting in cost of replacement; and a decrease in the efficiency of activated carbon with increased organic contaminants are limitations in the use of activated carbon  . Formation of byproducts like peroxides, ozonides, organobromine and bromate is associated with the use of ozonization  . A natural zeolite like clipnotilolite is high in cation exchange capacity due to its net negative charge on the outer surface    . When a natural zeolite is fortified with an overall positive charge using a surfactant, its affinity for cation changes for anion and it entraps negatively charged organic ions such as chromate and hydrophobic organic ions   . Clinoptilolite has high affinity for chromate and selenium and organic hydrophobics  and also for Pb2+  when modified by hexadecyltrimethylammonium bromide (HDTMA-Br). These unique attributes of a zeolite are both utilized in this study as we seek to alleviate pesticide residues in surface waters across Louisiana. This article reports the evaluation of natural zeolite, and HDTMA-Cl surface-mod- ified-zeolite on pesticide residue alleviation in surface water.
2. Materials and Methods
2.1. Water Sample Storage and Preparation
Ten surface water samples were collected from different locations in Louisiana. These were sourced from the pool of samples from the Pesticide Laboratory of the Agricultural Chemistry department, Louisiana State University through the Louisiana State Department of Agriculture and Forestry (LDAF). Water samples and their sources were as listed in Table 1. Each sample was labeled after its
Table 1. Water samples
*WM = Water monitoring.
source by abbreviating the name of the source. For instance, sample BPH was obtained from Bayou Pierre, Hwy 1 S of Powha. All water samples were stored at 4˚C until each was analyzed.
2.2. Pesticide Residue Extraction in Fresh Surface Water
The method used for pesticide residue extraction in surface water is same as earlier described  . Analyte samples, positive and negative controls were prepared in 2 replicates in vials and analyzed.
2.3. Water Filtration through Natural Zeolite-Clinoptilolite
Ten water samples―BPH, CLC, CDG, BBH, BRH, BCH, LBT, TRH, BPI, and BDC, were selected from the original pool of 35 samples of surface water studied for detection of pesticide residues. The criterion used in selecting those 10 samples was the water samples that had the most pesticide residues based on the results from a similar study  . As shown in Figure 1, each of the 3 compartments of the water filtration system used to filter surface water samples from top to bottom contained 20 g each, of gravel, sand and Zeolite. A funnel was placed on the topmost column and filtration was initiated. The filtrate was collected into a 1 liter amber color bottle as shown in Figure 1(b). For each water sample, a total of 1000 ml was filtered per 20 g of zeolite after which the filtration system was dismantled, cleaned by hot wash in soap, rinsed in running potable water three times and allowed to dry before re-assembled and re-used. Fresh zeolite was used for each sample.
2.4. Preparation of HDTMA-Cl (Hexadecyltrimethylammonium Chloride)-Surface-Modified Zeolite
As earlier described  , 0.056 M surfactant (HDTMA-Cl) was prepared to treat the natural zeolite. On a weighing balance, 1.43 g HDTMA-Cl was measured and poured into a 125 ml beaker containing 70 ml of milliQ water. With a gentle
(a) (b) (c)
Figure 1.(a) Water filtration system. ((b) & (c)) Surface water filtration through zeolite and HDTMA-Cl SMZ.
swirl until all surfactant dissolved, solution was poured into 100 ml graduated cylinder and milliQ water added up to 80.5 ml final volume. Twenty gram of natural zeolite was dispersed in the 80.5 ml of 0.056 M surfactant for 2 hours. The supernatant was drained away after 2 hours and the surface-modified zeolite (SMZ) was spread out on a clean aluminum foil to air dry overnight.
2.5. Water Filtration through HDTMA-Cl-Surface-Modified Zeolite
The water sample BRH was selected based on its highest volume of pesticide residue content. The filtration system for SMZ consisted of 3 columns in layers. The upper layer was empty followed by a middle layer of natural zeolite and base layer of HDTMA-Cl-SMZ.
2.6. Pesticide Residue Extractions in Both Zeolite-Filtered, and SMZ-Filtered Waters
As listed in Table 2, ten zeolite-filtered water samples were extracted for pesticide residues. The same extraction method used in pesticide residue extraction in fresh surface water as earlier stated was repeated for both sets of samples-10 zeolite-filtered samples and 1 SMZ-filtered sample. In each case, the same volume of 1000 ml of water was ran through the natural zeolite and the SMZ accordingly. Sample vials for the GC-MS were prepared and analyzed.
2.7. Gas Chromatography-Mass Spectrometry
GC-MS analysis was the same as earlier described  . On column concentration of the samples were calculated. Spiking rate was computed and the efficiency of the methodology was confirmed through the value of the spike recovery. On column concentration expected from a spike was calculated and it provided a clue to where matrix standard needed to be in order to use it to calculate the recovery rate. The amount of sample represented in the liquid injected onto column was calculated. Parameters obtained regarding the quantitation and retention time are as outlined in Table 3.
2.8. Statistical Analysis
Statistical analytical system (SAS) was employed to run paired student T-test in
Table 2. Effect of zeolite treatment on pesticide residue (ppb) in surface water.
Origin = original pesticide residue in surface water; Reduction = reduced amount of pesticide residue in surface water; Pesti resid = pesticide residue; Desethatz = Desethylatrazine; before = pesticide residue (ppb) in water sample before filtration through zeolite; 1st and 2nd = First and second pesticide residue quantitation reading after filtration through zeolite; PR = pesticide residue; ND = non-detected.
Table 3. Retention time and quantitation ion for target compounds and their degradation products.
λ = lambda; DesIsopropylAtz = desisopropylatrazine; MB46136fm = MB46136, Fip. met.; MB45950 = MB45950, Fip. met. Pendameth = Pendamethalin; Propicon2 = Propiconazole 2; Propicon1 = Propiconazole 1; λ-cyhalot1 = Lambda-cyhalothrin 1; λ-cyhalot = Lambda-cyhalothrin; Cypermethrin 1 = Cypermet1; Cypermethrin 2 = Cypermet2; Esfenvalerate 1 = Esfenvalera1.
order to compare the concentration of pesticide residues in the water samples before and after zeolite treatments. The alpha value was set at P = 0.05. That is, when the calculated P-value is less than 0.05 then a statistical difference can be declared.
3.1. Role of Natural Zeolite in Pesticide Alleviation
Reduction in pesticide residues was observed in the 10 zeolite-filtered surface waters analyzed (Table 2).
As shown in Table 2, reduction in pesticide residue levels following zeolite filtration ranged from the minimum of 10.9% to a maximum of 100%. Minimum reduction was recorded in metolachlor in sample BRH, while the maximum were in atrazine in samples BPH, CDG and LBT; metolachlor in samples CLC, LBT, BCH, TRH2 and BPI; bifenthrin in sample CLC; acetolachlor in BBH and BCH; azoxystrobin in BBH; desethylatrazine in BCH and BPI; metribuzin in BCH, TRH2 and BPI; and both clomazone and bromacil in sample BDC. A high reduction rate of 99.1% was found in atrazine in the same sample CLC. Atrazine was also alleviated in sample BRH up to 89.7%. Moderately high rates of reductions were also found in atrazine at the rate of 77.6% in sample BCH; 57.9% in sample TRH2; and 69.4% in sample BPI. Reductions recorded in sample BRH included 50% metribuzin and 62.5% propanil; and in sample TRH2 was 66.7% azoxystrobin.
Statistics of the means comparison for the pesticide residues found before and after filtering water through natural zeolite using a paired student t-test is as shown in Table 4 at Pcritical = 0.05. The difference between the atrazine levels before and after filtration of water sample through natural zeolite was highly significant in CLC and was significant in BRH and BCH. Difference between the atrazine levels was not significant in samples BBH and TRH2. Second analysis of BPI atrazine was not detected and therefore paired t-test not applicable. No statistical difference was found between the desethylatrazine levels before and after zeolite treatment in samples CLC and BRH and LBT. No significant difference was found before and after treatment with zeolite for the pesticide levels in metolachlor in samples BPH, CDG and BBH. There was no statistical difference between desethylatrazine levels before and after zeolite treatment in CLC, BRH and LBT.
3.2. Role of Surfactant-Modified-Zeolite (SMZ) in Pesticide Alleviation
As summarized in Table 5, following SMZ treatment of sample BRH, 6 pesticides were detected out of 8. Propanil and dimethenamid were undetected after SMZ treatment.
Greater reduction of pesticide residues was recorded (Table 6) in the sample BRH that was filtered through the surfactant-modified-zeolite (SMZ). A 50% reduction was observed as 4 out of the 8 residues found were reduced following
Table 4. Paired t-test comparison of pesticide residue means before and after zeolite treatment.
Sig. = Significance; NS = no significant difference found among the pesticide residue levels recorded before and after treatment with natural zeolite clinoptilolite; * & *** = significant difference and highly significant difference respectively found among the pesticide residue levels recorded before and after treatment with natural zeolite clinoptilolite; SD = standard deviation; SE = standard error; PR = pesticide residue; df (degree of freedom) = 1; Pr > |t| = calculated P value by SAS; Alpha = 0.05 (critical P value); NA = not applicable.
Table 5. Effect of surfactant-modified-zeolite (SMZ) on pesticide residue in surface water.
SD = standard deviation; ND = not detected.
Table 6. Percentage reduction of the pesticide residue (ppb) in surface water filtered through SMZ.
BZ = before zeolite treatment; AZ = after zeolite treatment; ASMZ = after surface-modified-zeolite; SMZ = surface-modified-zeolite; PR = pesticide residue.
filtration through SMZ. The 4 compounds that were reduced by SMZ compared to filtration through natural zeolite included atrazine @ 95% compared to 89.7% reduction with natural zeolite (NZ); 59.2% clomazone compared with 35.8% with NZ; 47.7% metolachlor compared with 10.9% with NZ and 50% metalaxyl compared with 25% with NZ.
As outlined in Table 7, paired t-test means comparison of pesticide residue before and after SMZ treatment was conducted. A significant difference (Pcalc = 0.003) was found in atrazine between the pesticide level recorded before and after the SMZ treatment of sample BRH. A highly significant difference (Pcalc < 0.0001) was similarly found in metalaxyl levels before and after SMZ treatment. In pesticide levels recorded for clomazone, desethylatrazine, metribuzin and metolachlor, there was no statistical difference found among them.
Further paired t-test comparison of pesticide levels was conducted between the levels recorded after treatment with natural zeolite and the levels recorded after treatment with surfactant-modified-zeolite. The outcome of this as outlined in Table 8 showed a statistical difference in metalaxyl, and the difference ob-
Table 7. Paired t-test comparison of pesticide residue means before and after SMZ treatment.
*Sig. = Significance; NS = no significant difference found among the pesticide residue levels recorded before and after treatment with Hexa decyl trimethyl chloride surfactant-modified-zeolite clinoptilolite; ** = very significant difference found between the pesticide residue levels recorded before and after treatment with HDTM-Cl SMZ; SD = standard deviation; SE = standard error; PR = pesticide residue; df (degree of freedom) = 1; Pr > |t| = calculated p value by SAS; Alpha = 0.05 (critical p value).
Table 8. Paired t-test comparison of levels of PR (ppb) of zeolite-treated and SMZ-treated sample.
Sig. = Significance; NS = no significant difference found among the pesticide residue levels recorded between zeolite treated and SMZ treated sample BRH; *** = highly suignificant difference found among the pesticide residue levels recorded between zeolite treated and SMZ treated sample BRH; SD = standard deviation; SE = standard error; PR = pesticide residue; df = degree of freedom; Pr > |t| = calculated p value by SAS; Alpha = 0.05 (critical p value).
served was highly significant (Pcalc < 0.0001). No statistical difference between treatment was observed for atrazine, clomazone, desethylatrazine, metribuzin and metolachlor. However, negative mean value and t value computed for desethylatrazine and metribuzin showed a negative trend because the levels recorded after filtration through the SMZ was higher than the levels after filtration through the natural zeolite. As outlined in Table 6, after filtration through natural zeolite desethylatrazine level was reduced from original 0.62 ppb to 0.38 ppb compared to 0.42 ppb for SMZ. After filtration through zeolite, metribuzin level was reduced from 0.34 to 0.17 compared to 0.23 ppb recorded after filtration through SMZ.
As obtained in this study, adsorption of metalaxyl using zeolite has been earlier reported  . Reduction in atrazine recorded in this study is similar to two reports earlier published  ,  , even though they used SDBAC (stearyldimethylbenzylammoniumchloride) as surfactant to modify the zeolite and HDTMA -Cl (hexadecyltrimethylammonium chloride) was used in this study as modifying surfactant. Further reduction of atrazine, clomazone, metolachlor and metalaxyl after filtration through SMZ conforms to the theoretical principle of effect of exchanging CEC (cation exchange capacity) property of clinoptilolite with an anion exchange capacity, thereby enhancing its ability to retain negatively charged organic ions that ordinarily would have escaped. Differences recorded in the pH (ranging from 6.8 through 7.7) of the surface water samples may have impacted the cation exchange capacity of the zeolite. This finding is in agreement with the result of a similar study  where they confirmed that the success of clinoptilolite in removing organic contaminations is a function of pH, temperature, contact duration, and initial concentrations of humic acid and ammonia. Pesticide residues were alleviated in all the samples whose pH ranged between 7.1 through 7.7, while the low pH 6.8 in sample BDC might be responsible for those pesticide residues that were non-detected. Similar to the assertion  that the optimum temperature at which zeolite could reduce organic contaminants in water is about room temperature which was the reason while sample waters were always allowed to acclimatize to room temperature after retrieved from cold storage. Findings in this study may also imply that water samples need to be above neutral pH in order for the zeolite to work at its optimum as earlier suggested  that sample water needs to be about the pH of natural water (pH 7.0) for the detection of residues to be at its best. High atrazine level of 6.2 ppb originally detected in the BRH sample which was reduced to 0.64 ppb when treated with zeolite, and further reduced to 0.31 ppb when treated with SMZ confirmed the adsorption capacities of the natural zeolite (clinoptilolite) and HDTMA-Cl SMZ as a potential remedy to the concentrations of this herbicide in surface water. Some residues were not detected in the water samples and this could be due to their low concentration probably tending toward infinitesimal amount as GC-MS detection limits were surpassed. It could also mean that they have been totally removed from the sample by the SMZ treatment. As opposed to the expected event that enhanced reduction be observed when filtered through SMZ, a reversed trend observed in desethylatrazine and metribuzin may imply that they have greater affinity for natural zeolite than for zeolite modified by HDTMA-Cl surfactant. Great affinities of clinoptilolite zeolite for ammonium ion  and that of SMZ for chromate and selenite  are an indication that any trace amount of NH4+, chromate or selenate in any of the 10 samples studied in this section may have been reduced. However, lack of measurement of these ions limited us from any information regarding this aspect. Part of future work would be to examine water for metal contaminants like arsenic   ; Fe and Mn  ; Cd and Pb  ; Pb2+  , and the cation NH4+.
Results obtained in the reduction of atrazine, metolachlor, bifenthrin, clomazone, desethylatrzine, metribuzin, propanil and metalaxyl are good to build upon as modern scientists aspire to provide a permanent solution to pesticide residues in surface water. This could be a basis for a large scale pesticide reduction in other forms of water like ground water and potable water as time goes on. Development of an industrial scale filtration system that could utilize zeolite as water filtration medium will be required in order to put the results obtained in this study into the effective use that will impact communities, national and international boundaries. Simplicity of this method with its low cost filtration system, coupled with the fact that it is free of any form of health risk will enhance its practical use and eventually lead to a global adoption of this methodology.
We appreciate the Louisiana State Department of Agriculture and Forestry for the support of this study.