The Dispersive Liquid–Liquid Extraction Method Coupled with HPLC and its Application in Determining S-triazine Group of Herbicides in Soil Samples

The dispersive liquid-liquid extraction (DLLE) is an environmentally benign process, which is based on simple, sensitive and rapid sample pre-treatment technique, coupled with HPLC-UV. This unique method has been designed for the separation of s-triazine as a herbicide residue in environmental soil samples. The influencing parameters which have been used to optimize the extraction efficiency include type, extraction solvent (ES) volume, dispersive solvents (DS), extraction time (mainly centrifugation time), pH, ionic strength for the addition of different salts. Firstly, tetrachloroethane was taken as exreaction solvent (ES) to extract pesticide residues from target samples. Furthermore, acetonitrile acted as dispersive solvent in the DLLE method. The value of linearity that has been reported with concentration range of 0.05-200 μgL -1 , and value of correlation coefficient (r) lies in between 0.9997. The recovery of the herbicide from three soil samples spiked between 20 and 100 μgL -1 were in the recovery in between the range 88.02% to 95.90.0% and the relative standard deviations (RSDs) were 2.7%. Limit of quantitations (LOQs) obtained in this method was 0.09 μgL -1 . These results significantly revealed that DLLE is a very accurate and reliable method to estimate the desired pesticide, even at trace amounts, in soil samples.


INTRODUCTION
Some common agro-chemicals extensively applied in agricultural field for the improvment of production of vegetables, result in great environmental concerns. The spread of problem is serious as over insecticide, fungicide and herbicide applied in the field of agriculture and may found at far distance affecting the non-target species such as air, water, soil and vegetable [1][2][3] . Development of resistance to pest clean of unwated grass in the timee of cultivation is another big problem. Now several new pesticides have been generated or greater dose of pesticide is administrated to counteract the pest resistance. In order to protect the lives on earth and to maintain the ecological balance, pesticide application must be regulated on utmost necessary, at least to control misuse. Now, determination of pesticide in water, soil and vegetable matrices proceeds in two-step processes like transfer of solute from parent solution to some desirable phase enrichment of solute to reach the desired level of detection limit and the analyte was transfered from aqueous phase to highly densed extraction solvent phase followed by its determination.
Among the different new techniques, liquid-liquid extraction is widely used. This is based on the well-known Nerst distribution law. Liquid phase microextraction (LPME) is modified method of LLE which required mainly high consumption of extraction solvent. The LPME can be classified into different categories like SDME (Single Drop Micro Extraction) 19 , HFLPME (Hollow Fiber Liquid Phase Micro Extraction) 20 , DLLME 21-23 and cloud point extraction (CPE) 24 . Firstly, The DLLME 25,26 was reported by Rezaee et al., in 2006 27 . This DLLME method is operationly very quick, easy, low cost and the enrichment factor is very high. Besides, compared to the reported methods, this method requires very small amoint of dispersive as well as extraction solvents, the time required for equilibration is short and the extraction efficiency is significantly high. These factors raise the novelity and greater applicability of the current DLLE method. An important solvent system (ternary component) has been applied in this current method for analyzation of trace amount of herbicide in soil. The present study focuses on the assess of DLLE suitability and its application to determine herbicide in environmental soil samples. To conduct these present experimental studies, the samples were collected from different layers near to the Jalangi River at Mayapur, India because the major section of the riverbank is extensively used for agricultural purposes, and the application of plant protecting chemicals i.e pesticides are also very high in amount and excess pesticides are sedimented in the soil through the flow of water. In this way, the quantity of pesticide contaminated soil near the river is also very high. The author believes that this study discussing about the usages of DLLME method to determine herbicide in the tergate samples is firstly reported. In this report, effects of different experimental factors, for example, variety as well as volume of extraction and dispersive solvents, time of extraction, and effect of salt, have been discussed.

Reagents and standards solutions
The herbicide, atrazine (CAS no 1912-24-9) was obtained from Sigma Aldrich (USA). Solvents used viz. CCl 4 , CH 2 Cl 2 , CHCl 3 , C 2 H 2 Cl 4 , CH 3 CN, CH 3 OH and CH 3 COCH 3 , C 4 H 8 O were of HPLC grade (Merck, India), rest of commercially available and highly pure reagents were used to carry out the experiments. 0.01(N) HCl or NaOH were used for maintaining the pH of the solution. The purification of water was processed by using a Milli-Q system (Millipore, Bedford, MA, USA). Solvents were filtered using 0.45 μm membrane filter to filtrate colloidal particles and inorganic elements for HPLC sample preparation. This type of membrane filter was also applied to filtrate supernate i.e acetonitrile extract and an aliquot of that extract was subjected to the DLLE method.

Preparation of Sample
Samples (soils) used to perform the experiments, were accumulated from the agricultural field near to the bank of the Jalangi river at Mayapur near (latitude: 23.4250146 and longitude 88.3906705), Nadia, West Bengal, India ( Fig. 1, marked in pink coloured circle on the map). Mayapur is situated adjacent to Nabadwip West bengal, at the confluence of two rivers, Jalangi and Bhagirathi. At first soil were dried under air, pulverized and sieved to sequentially grain size 250 µm. Accurately 1.0 g soil samples were weighed and shaken at 120 spm for 1 h in a 250 mL conical flux using 200 mL mixture of acetonitrile (1%) and Q-Millipore water. The solutions used in this experiment were centrifuged at the speed of 800 rpm for 15 min and subsequently filtered by using 0.45μm membrane. Thus, soil samples containg large amounts of small diameter colloidal particles terbidity and leaching concentration of inorganic hazourdous elements were filtered during preparation of stock solutions.

Instrumentation
The chromatographic analysis was performed on Cecil (model CE 4201PC) HPLC coupled with Shimadzu UV-Vis recording spectrophotometer using manual injector having capacity 20 μL. The analytes were separated on Hyper-clone 5μ ODS (C18), 120A (150 X 4.60m: particle size 5μ) fitted with quaternary pump were applied to separate the analytes using acetonitrile: water [(90:10,v/v)] as a mobile phase at flow rate of 1.0 mL.min -1 . The column temperature and the detector wavelength (λmax) were adjusted at 30°C and 220 nm respectively. The volume of the solvent used for injection was 20 μL. Rotofix centrifuge was used for phase separation. A Systronics, India: model no 335 digital pH meters combined with glass electrode was also used.

Procedure of Dispersive Liquid-Liquid Extraction
5.0 mL of aqueous sample was transferred into a 10 ml glass test tube with screw cap and having conical bottom. To obtain the best extraction result, dispersive solvent (DS) acetonitrile (0.1 mL) was used, which was homogenised with a 0.9 mL of tetrachloroethane as extraction solvent (ES) and this aliquot was quickly added to the sample solution using a dispovan medical syringe. On gentle shaking, a hazy solution appeared first forming fine droplets. To obtain the extreme extraction, this mixture was centrifuged at the speed of 4000 rpm for 5 minute. The fine droplets were allowed to settle, and the upper phase was withdrawn using a micro syringe. The sediment phase diluted with acetonitrile, was run through HPLC-UV for solute characterization and subsequent spectrophotometric determination.

Mathematical Representation: Enrichment Factor (EF) and Recovery Percent (RP)
To explain the effects of experimental parameters, enrichment factor (EF) and extraction recovery (EP) the following equations were used; Where, C sed , V sed & W sed are concentration, volume and amount of solute in sediment phase, C o , V o & Wo represent concentration, volume and solute amount in aqueous phase. The equation (3) was obtained by combining euations (1) and (2), where EF and RP can be related as.

RESULTS AND DISCUSSION
DLLE miniaturized form of liquid-liquid extraction where mililiter volumes of extraction solvent (more dense solvent) were used for extraction of targate analytes from solution. The Distribution coefficient (K) was stated as the ratio of the analyte concentrations in extraction solvent and sample solution respectively. This DLLE method was suitable for analytes having moderate to high lipophilic character and not suitable for neutral highly hydrophobic analytes. A very crucial point to be noted here that the distribution coefficient of acidic and alkaline analytes were enhanced by maintaining pH of that analytes whereas analytes remained in nonionic form. In DLLE, solute remains in water solution. A couple of dispersive solvent and extraction solvent were injected with a dispovan syringe to the sample solution. The cloudy appearence (aqueous/ DS/ES) was seen and finally this dispersed droplets of ES sedimented at the bottom of the tube, were used for chromatographic detction. The extraction efficiency significantly depended on the types as well as volumes of extraction and dispersive solvents, the extraction time, pH and ionic strength of the sample solution. This study focused on the assess of DLLE suitability to determine pesticide as well as herbicide concentrations specially atrazine concentrations in soil samples. The parameters affecting the extraction efficiency were also optimized.

Choice of extraction solvent
In this efficient method, choice of extraction solvent was very important factor. The extraction solvent must have higher density than water, good chromatographic nature and extremely well to extract analytes. Based on these criteria, in this study four chlorinated solvents viz. CCl 4 (1.59 gmL -1 ), CH 2 Cl 2 (1.33 gmL -1 ), CHCl 3 (1.47 gmL -1 ) and C 2 H 2 Cl 4 (1.60 gmL -1 ) were examined and CCl 4 acted as non polar extraction solvent. In each experiment, 0.3 mL of DS and 0.7 mL of ES were injected. The effect of the extraction solvents on the recoveries using different combonation dispersive solvents was shown in Fig. 2. It was observed that C 2 H 2 Cl 4 in combination with acetonitrile provided the highest extraction efficiency. Therefore, C 2 H 2 Cl 4 was selected as an extraction solvent

Choice of dispersive solvent
The  Fig. 2. The combination of extraction solvents with dispersive solvents produced a two-phase system and this effect on extraction recovery was shown in Fig. 2. It was noticed that, acetonitrile gave the best extraction efficiency and therefore, acetonitrile was opted as the dispersive solvent for subsequent studies.

Influence of extraction solvent volume
It this study, volume of extraction solvent varied from 0.1mL to1.5 mL. When less than 0.1 mL volume of C 2 H 2 Cl 4 was used no sediment appeared. Analyte must have better extractability in ES, compared to the water i.e., ES must be hidrophobic in nature. The ternary phase formation is an important parameter for selection of ES. To observe the effect of ES volume on recovery of analyte, the volume varied in the range 0.1 mL to 1.5 mL for a fixed volume of acetonitrile (0.1 mL) as DS which was shown in Fig. 3. Recovery percentage increased with increase of ES volume for at fixed volume of DS. After reaching the maxima, the recovery percent decreased with the increase of ES volume.

Influence of dispersive solvent volume
It was minutely observed that when the volume of acetonitrile changed from 0.0 to 0.3 mL, the trend of recovery increased upto 0.1 mL then decreased Fig. 4. Therefore, it is definite that the minimum volume of DS for highest recovery is 0.1 mL for atrazine extraction. Such difference of extraction-dispersive volume for pesticide recovery may be explained from the difference in extractability arises due to solubility, dispersibility and cloud formation ability. At fixed volume of ES, percent recovery of presticide was lower and at higher volume of DS, it was found lower recovery percent. Without DS, the analyte could not disperse into ES properly and inhibited the formation of the cloudy solution resulting incomplete or poor extraction. At higher volume of DS, however, due to dilution effect extraction becomes less. Thus, the optimum volume of ES 0.9 mL and DS for atrazine extraction were 0.9 mL and 0.1 mL respectively. significant factor in this extraction process containing acidic or basic analytes. To investigate the effect of pH on the extraction efficiency of DLLE, we took a limit range 2 to12 in one pH interval. The sample solution was adjusted to such a pH that analytes extraction was very effective. The pH was adjusted with dropwise addition of 0.01(N) NaOH and HCl. It demonstrated that the proposed method significantly improved the detection sensitivity compared with other techniques were shown in Fig. 6. Results showed that the recovery of atrazine increases by increasing pH from 2 to 6.1, when mxture of CH 3 CN acts as disperser and C 2 H 2 Cl 4 extraction solvent give best perfomance. However, subsequent increase in pH led to a decrease in extraction efficiency. It was observed that pH of the solution was 6.1 and which was optimum for extraction.  Fig. 5. To accelerate the phase separation process, a centrifuge was used, and this process consumed less time. Unwanted long centrifugation was avoided because this might disturb the phase separation due to heat generation and might dissolve the anlyte.

Influence of salt addition
Addition of salt often ameliorates extraction in conventional LLE, because of the salting out effect. To evaluate the possibility of the salting out effect, extraction recovery of herbicide by DLLE was studied range of 0-12.5% (w/v) of salt. It was found that recovery increases in the order, (NH 4 ) 2 SO 4 >KNO 3 >KCl>NaCl. In this study, all experiments were performed repeatedly. (NH 4 ) 2 SO 4 shows better performance than other. Increasing (NH 4 ) 2 SO 4 amount more than 2.5%w/v causes a small decrease in the extraction recovery of atrazine shown in Fig. 7. Therefore, 2.5% (NH 4 ) 2 SO 4 (w/v) was optimum concentration used in this experiment. Each case the sample volume is 5.0 mL.

Quantitative analysis
The important parameters obtained from the experimental results, implied for the validity of this method. For test of efficiency of herbicide extraction a series of working solution were prepared. Each concentration level was repeatedly used in this extraction procedure. The linearity was range of 0.05-200μgL -1 . The correlation coefficient symbolized by 'r' = 0.9997. The observed LOD of this method was 0.03μgL -1 , the enrichment factors ranged above 600, with RSDs was from 2.7% and the recoveries ranged from 93.25% to 98.59% Table 1.
results of these recovery results obtained from soil samples were given in Table 2 Table 2. The metod provids good repeatibility and high relative recoveries of herbicide form environmental samples (soil samples) were above 88.0%. The Fig. 8. showed the chromatograms of atrazin in. The chromatographic analysis was performed on Cecil (model CE 4201PC) HPLC coupled with Shimadzu UV-Vis recording spectrophotometer.

Real sample analysis
Three soil samples were collected from three different layers near to the Jalangi River at Mayapur. The source of this river is Padma (Jalangi) and mouth is Bagirati (Mayapur) near about 250 km. It flows over the district Murshidabad and Nadia. The

CONCLUSION
In this work, the application of DLLE coupled with HPLC-UV was successfully discussed and applied for the extraction and determination of atrazine herbicide from soil samples. This simple and versatile method provides good enrichment factors and efficient separation and recoveries for the target analyte. In comparison to solvent extraction, it is much safer, since only a small volume of the solvent is used. Finally, experimental results clearly indicate that the DLLE method gave a swift and economical procedure for recovery of herbicide from environmental sample. The recovery greater than 91% compared with other methods shown Table. 3. Finally, this developed method was succesfully appleal for the recovery of pesticide from different matrices.