It has been developed a small-scale, simple, and rapid dispersive liquid-liquid microextraction (DLLME) procedure in combination with fiber optic-linear array detection spectrophotometry (FO-LADS) with charge-coupled device (CCD) detector benefiting from a micro-cell. The official reference methods (ASTM D2330 - 02, ISO 7875-1) which require tedious procedures were replaced with modified method, as a result, it has achieved a major reduction in sample size, elimination of the use of expensive glassware, and a decrease in the quantity of chloroform used as well as much more gain in sensitivity. Our presented method requires only one twentieth of sample (5.0 mL), less than one three-hundredth of microextraction solvent (chloroform = 138 µL), and much reduced in analytical time compared with official analytical methods (less than one minute). The calibration curve was linear in the range of 0.06 Ã- 10-1 - 0.8 Ã- 10-1 mg L−1 of sodium dodecyl sulfate (SDS) with a correlation coefficient (r) of better than 0.99 and the LOD was 0.02 Ã- 10-1 mg L−1. The repeatability of the proposed method (n=7) were found to be 4.5 and 3.6 % for the concentration of 0.03 and 0.07 mg L−1, respectively. The enrichment factor was found to be 75 for SDS.
Keywords: Dispersive liquid-liquid microextraction · Water analysis · Methylene blue active substance · Anionic surfactant · Fiber optic-linear array detection spectrophotometry
1. Introduction
A growing public concern over protecting our environment obligate chemists, including analytical chemists, to change their activities in such a way that they will be conducted in an environmentally friendly manner. Sampling, and especially sample preparation, frequently involves generation of large amounts of pollutants. This is why sample preparation techniques that use a small amount of organic solvent, or none at all, have been developed [1-4].
Anionic surfactants (AS) are widely used in household cleaners, industrial detergents and cosmetic formulations. The surfactants expelled to natural water reservoirs as municipal and industrial wastes are well known to have adverse effects on aquatic organisms; hence the monitoring of surfactants in environmental samples is of great importance [5, 6].
For the measurement of total surfactant concentration, titration methods have been extensively explored [7, 8]. Several ion-selective electrodes sensitive to anionic surfactants have been reported so far [9-11].
Anionic surfactants are usually determined by spectrophotometric methods using methylene blue (MB), this standard methods being used to determine AS in the surface and tap-water samples (ASTM D2330 - 02, ISO 7875-1) [12, 13]. The method is based on the formation of blue-coloured chloroform extractable ion-pair between the AS and the cationic MB. This requires three successive extractions of AS-MB content in 100 ml of sample with 15, 10, and 10 ml of chloroform. The ion-pair is determined by spectrophotometry, measuring the absorbance at 650 nm. However, these official methods are not only long and tedious but also require great quantities of sample and chloroform which has harmful effect on chemists and environment. Besides, this method needs lot of laboratory glassware, make these operations extremely expensive and uncomfortable for the operator. So it seems necessary to search for new offers as alternatives for the aforementioned method in order to increase the laboratory productivity, operator safety, comfort, and to reduce drastically the reagents consumption and waste production.
Koga et al. proposed a reduction of the size of sample employed for AS determination in water, being modified this method to use 50 ml of water and 5 ml chloroform, having obtained a six times increase of the laboratory productivity [14]. An other simplified methods that reduce the quantities of reagent by using a certain kind of adsorbent have been proposed [15]. However, this method also involves tedious procedures. Also other researchers studied primary biodegradation of AS in aerobic screening tests based on the formation of ion-pair of AS and MB [16].
By early 2006, Assadi and his research group innovated an attractive, high performance and powerful liquid-phase microextraction (LPME) method which named their techniques "Dispersive liquid-liquid microextraction" (DLLME) [17-19]. Beyond the trait of simplicity of operation and rapidity, consumption of microextraction solvent at the micro-level volume and compatibility with analytical instruments are other profitable features of DLLME as a sample pretreatment method [20-25]
For highly sensitive, accurate, rapid, and inexpensive measurement with consumption of extraction solvent at micro-level volume, we propose a simplification of the spectrophotometric MB method that can be useful for determining anionic surfactants in aqueous samples. A successive DLLME in combination with fiber optic-linear array detection spectrophotometry (FO-LADS) with charge-coupled device (CCD) detector benefiting from a micro-cell was used for this purpose.
2. Experimental
2.1 Reagent and standards
The reagents used in the experiments were of analytical grade: MB (used as a cationic dye), sodium dodecyl sulfate (SDS, employed as a representative anionic surfactant), acetone as disperser solvents, chloroform as microextraction solvent, NaOH, HNO3 (65 %), HCl (37 %), acetic acid, and sodium acetate for making buffer solution) and obtained from Merck (Darmstadt, Germany). Absolute ethanol (> 99.6 %) purchased from Bidestan company (Qazvin, Iran).
The required quantity of SDS was dissolved in pure water to make standard solution of 1000 mg L-1. The stock solutions of MB (3 Ã- 10-3 mol L-1) were prepared by dissolving appropriate amounts in double distilled water. All the plastic and glassware were cleaned by soaking for 24 h in 10% v/v HNO3. After cleaning, all containers were thoroughly rinsed three times with double distilled water and twice with acetone prior to use. No any detergent was used to clean glassware because it is difficult to remove from surfaces and causes high results.
2.2. Apparatus and Instrumentation setup
The fiber optic-linear array detection spectrophotometer was perched from Avantes (Eerbeek, Netherlands). The light beam from the UV-Vis source (Deuterium-Halogen) was focused to the sample micro-cell (Starna Scientific, Essex, England, Cat. NO. 16.40F-Q-10/Z15). The spectrograph accepts the light beam transmitted through the optical fiber and disperses it via a fixed grating across the 2048 element CCD-linear array detector. The instrumental parameters are listed in Table 1. A Universal EBA 20 centrifuge equipped with an angle rotor (Angle rotor for 8 Ã- 15 mL tubes, 6000 rpm, Cat. No. 2002) were obtained from Hettich (Kirchlengern, Germany). An adjustable pipette (10-100 µL) was prepared from Brand (Wertheim, Germany). All 0.1, 1.0 and 2.5 mL syringes were prepared from Hamilton (Reno, NV, USA).
To clean out the micro-cell, avoid any memory effect and improve the repeatability of procedure, it was washed three times by about 2 mL of acetone between each analysis and dried with a stream of cold air by use of a hair dryer.
2.3. Reference procedure
Hundred millilitre of sample was placed into a 250 ml separating funnel and 10 ml of a 1 Ã- 10−3 mol L-1 MB solution and 15 ml chloroform were added. After shaking the mixture vigorously for 1 min, the two phases were let to separate and chloroform layer taken for analysis. Each sample was extracted additionally two times using 10 ml portion of chloroform and absorbance measurements were made at 650 nm in front of an external calibration prepared from SDS. Solutions in the range between 0.1 and 0.5 mg L−1 were extracted in the same way than samples.
2.4 Recommended analytical procedure
Into a series of screw cap glass test tube with conical bottom 5.0 mL of the standard SDS solutions at the concentration in the studied range were pipetted out. Then 25 µL of 3 Ã- 10-3 mol L-1 MB standard solution was added. Afterwards, 2.00 mL ethanol (disperser solvent) containing 138 µL chloroform (microextraction solvent) was injected rapidly into the sample solution using a 2.50-mL syringe. This injection led to a cloudy solution, caused by the fine droplets of chloroform into the aqueous sample. The phase separation was accelerated by centrifuging at 5500 rpm for 3 min. After this step the dispersed fine droplets of chloroform were settled at the bottom of the aqueous solution in conical test tube. Subsequent to this procedure, for evacuating the upper aqueous solution a long needle connected to 10-mL injection syringe was immersed down in to test tube and pulled the plunger up till moment 200-300 µL of aqueous phase remains at the top of organic layer. The volume of the settled organic phase was determined using a 100-mL microsyringe at 25 °C which was 65±2 µL. Sixty micro-liter of this settled phase was removed by micropipette and introduced into micro-cell. The ordinary absorbance of AS-MB ion-pair in chloroform was measured at the wavelength of 650.0 nm by means of FO-LADS.
3. Result and discussion
In order to obtain a high sensitivity, the parameters affecting the DLLME such as the type of the microextraction and the disperser solvents as well as their volume, concentration of MB, pH, and the microextraction time were optimized.
The enrichment factor (EF) was defined as the ratio of the analyte concentration in the settled phase to the initial analyte concentration in the aqueous sample. The analyte concentration in the settled phase was calculated from the calibration graph obtained by the conventional liquid-liquid extraction (LLE)/FO-LADS (extraction conditions: 2.0 mL standard water sample in the concentration range of 4.5 Ã- 10-4 - 1.5 Ã- 10-3 mol L-1 of MB and 1.5 - 5.0 mg L-1 SDS which extracted with 2.0 mL chloroform).
3.1. Reaction of SDS and MB
The equilibrium between SDS, MB and the distribution of SDS-MB ion-pair in water and chloroform has been qualitatively reported in the literature [14]. The AS dissolved in water are slightly soluble in chloroform. On the other hand, MB dissolves well in both, chloroform and water, providing a blue color solution in all the cases. When pure water is mixed with a chloroform solution of MB, the blue color is rapidly transferred to the aqueous phase.
3.2. Effect of ion-pair formation condition parameters
The overall ion-pair formation condition of SDS and MB is concentration of each, pH as well as time needed. Our attempts were primarily centered on optimizing these parameters under our microextraction conditions (DLLME).
In this study the time required for ion-pair formation were tested between 0 sec -10 min. The results, deriving from the ion-pair formation using different reaction times, exhibited that the reaction time has no any effect on ion-pair formation efficiency and longer time period did not improve the reaction. In order to determine the optimal pH for the ion-pair formation, several sample pH values were varied from 2.5 - 7.5 to test the ion-pair formation of AS and MB in 5.0 mL water samples containing 0.04 mg L-1 SDS and excess amount of MB. The highest microextraction efficiency was achieved in the pH of studied range and we found that in the alkaline solution MB it self would extract into chloroform in absence of any MBAS. In optimization procedures no any buffer solution were used because after adding reagents the pH of solution become slightly acidity in desired range.
The influence of the MB concentration on the ion-pair formation/microextraction efficiency was performed in the range of 0 - 2.1 Ã- 10-5 mol L-1 while the concentration of SDS was 0.04 mg L-1. During the variation of this concentration the other experimental variables remained constant. The results demonstrated that by increasing the MB concentration up to 1.5 Ã- 10-5 mol L-1 the microextraction efficiency increased and, then, no variation were observed (as depicted in Fig. 1). Considering the fact that proposed method is linear up to 0.08 mg L-1, therefore, the amount of 5 Ã- 10-5 mol L-1 MB was selected as consider enough excess amounts.
3.3. Influence of the microextraction solvent kind and volume
The selections of an appropriate microextraction solvent have a high importance role to get a high sensitivity DLLME, so kind and volume of it were studied and optimized. Microextraction solvent should have special characteristics in DLLME; it should have very low solubility in water, extraction capability of interested compounds, and much density than water. Chloroform and carbon tetrachloride are available as the most famous microextraction solvents in DLLME. During our primary studies we found that carbon tetrachloride is not capable to extract the ion-pair of SDS-MB at all. Moreover, the recommended solvent in the standard methods is chloroform; therefore, it was our extinguished choice.
To investigate the effect of microextraction solvent volume, experiments were performed by using 2.00 mL ethanol containing different volumes of chloroform (138, 143, 148, 153, 158 and 163 mL). By increasing the volume of chloroform from 138 to 163 µL, the volume of the settled phase increases approximately from 65 to 90 µL. According to results (Fig. 2), absorbance decreases with increasing the volume of chloroform; it is clear that by increasing the volume of chloroform the volume of the settled phase increases. Subsequently, at low volume of the microextraction solvent high absorbance or enrichment factor was obtained.
3.4. Influence of the disperser solvent kind and volume
In DLLME, selecting an appropriate disperser solvent is important, since disperser solvent should be miscible with both microextraction solvent and aqueous sample. For the sake of acquiring the most suitable disperser solvent, two kinds of rather safe disperser solvents: acetone and ethanol were studied. A series of sample solutions were studied by using 2.00 mL of each disperser solvent containing 138 µL of chloroform and the enrichment factors were investigated. The results showed that ethanol showed much better efficiency than acetone (enrichment factor of 75 and 17, respectively). Less toxicity and the higher microextraction efficiency of ethanol make it a better choice.
After choosing ethanol as disperser solvent, it is necessary to optimize the volume of it. The influence of the disperser solvent (ethanol) volume on the microextraction efficiency was tested over the range of 0.50 - 2.00 mL, but the variation of the ethanol volume (disperser solvent) caused changes in the settled phase volume. Hence, it was impossible to consider independently the influence of the ethanol volume on the microextraction efficiency in DLLME. To avoid this problem and in order to attain a constant volume of the setteled phase, the ethanol and chloroform volumes were changed simultaneously. The experimental conditions were fixed and included the use of different ethanol volumes: 0.50, 1.00, 1.50, and 2.00 mL, containing 97, 102, 121, and 138 μL of chloroform, respectively. Under these conditions, the settled phase volume remained constant (65 ± 2 μL). Fig. 3 shows the curves for absorbance of SDS-MB ion-pair versus the volume of ethanol. The absorbance increased, when the ethanol volume increased from 0.50 to 2.00 mL of ethanol as disperser solvent. According to the results, a 2.00 mL ethanol was chosen as the optimum disperser solvent volume.
3.5. Influence of the microextraction time
Microextraction time (interval time between the injection of a mixture of disperser solvent and microextraction solvent, before starting to centrifuge) is important factor that may be effects microextraction efficiency of analytes from aqueous phase to organic phase. The variation for microextraction efficiency of SDS-MB as a function of microextraction time was studied in the range of 5 sec - 10 min. The resulting data, displaying that the microextraction time has no significant effect on the microextraction efficiency for the target compound. It was revealed that after the formation of the cloudy solution, the contact area between the microextraction solvent and the aqueous phase was considerably large, delineating why the extraction equilibrium could be established very fast. In this method the most time-consuming procedure was centrifugation of the sample solution in the microextraction procedure, which was about 3 min. Considering the fact this period of time (3 min) is for eight test tube (microextraction vessels), the time required for handling one test tube is less than 25 seconds.
3.6. Analytical characteristics of the method
To evaluate the practical applicability of the proposed DLLME/FO-LADS technique for determination of MBAS in water samples, several analytical performance characteristics such as enrichment factor, linearity, limit of detection (LOD) and repeatability were investigated under optimized conditions. The calibration curve was linear in the range of 0.06 Ã- 10-1 - 0.8 Ã- 10-1 mg L−1 of SDS with a correlation coefficient (r) of better than 0.99.
The LOD, defined as CL =3 SB/m (where CL, SB and m are the limit of detection, standard deviation of the blank and slope of the calibration graph, respectively), was 0.02 Ã- 10-1 mg L−1. The repeatability of the proposed method expressed as relative standard deviations (RSDs, n=7) were found to be 4.5 and 3.6 % for the concentration of 0.03 and 0.07 mg L−1, respectively. The enrichment factor was found to be 75 for SDS.
3.7. Effect of diverse ions and application to practical samples
Any organic or inorganic compound that will form a chloroform extractible ion-pair with MB will interfere by producing high results. These positive interference include organic sulfonates, carboxylates, phosphates, and phenols, as well as inorganic cyanates, chlorides, nitrates, and thiocyanates. On the other hand, any compound effectively competing with MB to form an AS ion-pair will give negative results. These negative interferences cause by some amines and have analytical significance in the case of quaternary ammonium compounds. For pretreatment of MBAS in all waters and waste waters that contain interfering substances the following procedure is recommended by ASTM reference method. The selected sample is hydrolysed by boiling under partial reflux with hydrochloric acid. The residual products are neutralized to a controlled pH value, and reacted with 1-methylheptylamine. The resulting ion-pairs are extracted into a chloroform phase and evaporated to dryness on a steam bath. The amine component of the ion-pair is removed by boiling in an aqueous alkaline media and the isolated MBAS are then determined under the described reference procedure. Also other researchers examined the effect of various diverse ions on the determination of AS by similar method [14, 6].
In order to establish the validity and applicability of proposed method, it was applied to the determination of AS in several real water samples (mineral, tap, and well water samples) by proposed method. For this purpose, 5.0 mL of each sample was preconcentrated using DLLME technique as described before (pH was adjusted with acetic acid/sodium acetate buffer if necessary). In order to assess matrix effect, the standard addition method was applied for the determination of AS (at spiking levels of 0.02 and 0.05 mg L-1) in spiked real samples which the relative recoveries of analytes are mentioned in Table 2. The obtained results were compared with those obtained from spiked distilled water. In all cases, the spike recoveries confirm the reliability of the proposed method. The obtained relative recoveries indicates that matrix does not influence the microextraction efficiency in the mentioned samples (no serious interferences), therefore, there was not any obligation to remove interferences. As it can be seen in table 3, the performance of proposed method shows distinct advantages over other methods with reference to sample volume, extraction solvent volume, RSDs, LODs and linear dynamic range
Conclusions
This study demonstrated that DLLME procedure with very pleasant and robust characteristics for assay of AS seems to offer potential candidates for reference method, which utilizes very small amount of microextraction solvent as well as its low cost. Moreover, newly DLLME procedure in combination with FO-LADS equipped with charge-coupled device (CCD) detector benefiting from a micro-cell demonstrated that LPME (DLLME) could be combine with spectrophotometer system despite of micro-level sample volume without any dilution and decreasing the sensitivity. Analysis of several real samples for AS content illustrated the accuracy, reliability, simplicity, reliability and cheapness of method. It appears to be a time-saving technique, mainly for laboratories performing analysis of a large number of samples with a rapid reporting time. Also we suggest the applicability of this method for monitoring the biodegradation of AS.
Acknowledgement
The authors would like to thank University of Tehran for financial support as grants.
