A GREEN VORTEX-ASSISTED IONIC LIQUID-BASED DISPERSIVE LIQUID–LIQUID MICROEXTRACTION METHOD FOR PRECONCENTRATION AND DETERMINATION OF TRACE CADMIUM IN FOOD SAMPLES

Objective: Green, easy, and sensitive vortex-assisted ionic liquid-based dispersive liquid–liquid microextraction technique (VA-IL-DLLME) was developed to preconcentrate and determine trace quantities of cadmium (Cd2+) ions from real food samples, before detection by flame atomic absorption spectrometry. Methods: The proposed technique base on the utilization of IL (1-hexyl-3-methylimidazolium tris(pentafluoroethyl)trifluorophosphate) as an extraction solvent for Cd2+ ions after the complexation with 2-(2’-benzothiazolylazo) chromotropic acid at pH 8.0. The impact of different analytical parameters on microextraction efficiency was investigated. The validation of the proposed procedure was verified by the test of certified reference material (SRM spinach leaves 1570A) applying the standard addition method. Results: In the range of 1.0–300 μg/L, the calibration graph was linear. Limit of detection, preconcentration factor and the relative standard deviation (RSD%, 25, 150, and 250 μg/L, n=5) were 0.2 μg/L, 100, and 2.0–3.2%, respectively. Conclusion: Green, VA-IL-DLLME method was developed and applied to preconcentrate and determine of trace quantities of Cd2+ in real food samples with satisfactory results. The obtained recovery values showed good agreement with the certified values.


INTRODUCTION
Cadmium (Cd 2+ ) is considered to be one of the important hazardous heavy metals due to its toxicity in living organisms at low concentration levels [1]. Cadmium was categorized as a carcinogen compound to humans and a wide variety of chronic problems in liver, kidneys, renal, and lungs were caused by high exposure to cadmium [2]. Therefore, the estimation of trace Cd 2+ in different types of various samples is a very important objective for chemists using effective methods [3][4][5]. Trace quantities of Cd 2+ in real samples are less than the detection limit of some instruments such as flame atomic absorption spectrometry (FAAS) or GF-AAS and coexisting ion interference is two limitations in estimations [6][7][8].
The main of the proposed work was to develop a green, innovative VA-IL-DLLME technique in combination with FAAS for preconcentration and accurate determination of trace quantities of Cd 2+ in real food samples. In the proposed procedure, hydrophobic IL,(1-hexyl-3-methylimidazolium tris(pentafluoroethyl)trifluorophosphate [HMIM][FAP]) as the extraction solvent and 2-(2'-Benzothiazolylazo) chromotropic acid (BTANC), as complexing agent was selected. Vortex-assisted was utilized to accelerate the separation and preconcentration step. The method validity was examined by the analysis of certified reference materials.

Instrumentation
Atomic absorption spectrophotometer (Agilent 55 AA; Agilent Technologies Inc., Santa Clara, USA) was utilized to determine the analyte metal. The instrumental conditions were set in accordance with the instructions of the manufacturer. The microinjection method was utilized to introduce the sample to the nebulizer of the FAAS [26].
An Adwa AD1000 pH-meter (Szeged, Hungary) was utilized to measure the pH-values of prepared buffer solutions. Vortex mixer (iSWixVT, Neuation Technologies Pvt. Ltd., India) was used to mix thoroughly the sample solutions and to assist the extraction of analyte from sample matrices. A centrifuge (Isolab, Germany) was utilized to accelerate the separation of analyte from the sample solution. Milli-Q was utilized to obtain deionized/bidistilled water (Millipore, USA). Glasswares were kept in nitric acid (5.0% v/v) overnight, rinsed and cleaned many times with bidistilled water before the experiment.

Reagents and solutions
All reagents and chemicals used were of high purity and acquired from Merck (Darmstadt, Germany) and Sigma-Aldrich (St. Louis, USA) The proposed procedure was successfully applied to fruit juices samples "orange, apple, and grape juices," acquired from the local market, Egypt. All samples are filtered through a cellulose membrane filter of 0.45 μm pore size (Millipore Corporation, Bedford, MA, USA) to eliminate hanging materials before being acidified with diluted HNO 3 and stored at 4°C. To oxidize the organic content of fruit samples, H 2 O 2 (30% w/v) and HNO 3 (65% m/m) were utilized. The mixture was refluxed for 1.0 h approximately at 100°C on a hot plate and then the obtained solution was evaporated to approximate dryness. The residue was dissolved in water and diluted to 50 mL with deionized water and an acetate buffer solution was added to adjust the sample pH to 8.0. Then, the preconcentration procedure was completed as previously described. The concentrations of Cd 2+ ions were evaluated by FAAS.

Food samples
The different vegetable samples (cabbage, spinach, and tomato) were gained from the local markets of Egypt. For 24 h in an electric oven, the food samples are dried at 80°C and homogenized by grinding in an agate mortar. For a wet acid digestion procedure, SRM 1570A spinach leaves, and food samples (0.2 g) were treated with 2.0 mL of a mixture of concentrated HNO 3 -H 2 O 2 (2:1, v/v) into the beaker and kept at room temperature for 10 min, then heated on an electric hot plate until semidried mass was obtained. Digested semi-mass samples were dissolved in 5.0 mL of deionized water and filtered through a cellulose membrane filter of 0.45 μm pore size [5]. Then, the developed VA-IL-DLLME procedure was satisfactorily applied to the samples.

Influence of pH
The pH has a crucial impact on the extraction yield and the formation of the metal-chelate complex [27][28][29][30]. Consequently, the influence of pH on the VA-IL-DLPME microextraction technique of Cd 2+ -BTANC complex was studied at a pH range of 3.0-10. At pH 7.5-8.5 with phosphate buffer solution as illustrated in Fig. 1, the extraction recovery of Cd 2+ increased with increasing pH and maximum quantitative values are accomplished. In subsequent studies, 4.0 mL of buffer solution (pH 8.0) was chosen.

Influence of amount of BTANC
The BTANC amount has an important influence on the extraction recovery of Cd 2+ to give quantitative results. Various concentrations of BTANC were examined within the range of 0.05-0.5% (w/v), and results curvature is shown in Fig. 2. The recovery was increased by increasing the BTANC concentration up to 0.2% (w/v) and higher amounts of BTANC have no significant effect on the extraction recovery. Therefore, 0.2% (w/v) of BTANC was utilized as the optimum amount for total complexation and maximum extraction recovery in further studies.

Influence of IL
In the microextraction technique, choice of the type and volume of IL were important which has a large influence on the Cd 2+ extraction efficiency. In the present work, [HMIM][FAP] was selected as the extraction solvent. This IL has some characteristics such as; its good thermal stability, hydrophobicity, and negligible vapor pressure. Consequently, the volume of IL was studied in the range of 50-400 µL (Fig. 3). The extraction competence of Cd 2+ was enhanced with IL volume in the range of 150-250 µL. At higher volume, no significant change in recovery was obtained. Hence, 200 µL of IL was chosen as an optimum volume for all the subsequent studies and to get a better enrichment factor (EF).

Influence of dispersive solvent type and volume
In the microextraction technique, choice of dispersive solvent is a significant parameter to form dispersed fine droplets of IL because it must be miscible in both the aqueous and the IL phases. Hence, various dispersive solvents such as (methanol, ethanol, acetonitrile, acetone, and THF) were utilized. The high recovery was observed using methanol as a dispersive solvent. The effect of ethanol volume of the analytical signal of Cd 2+ was tested in the range of 50-700 μL (Fig. 4). In subsequent studies, the suitable volume of methanol (500 μL) was selected as optimal and provided the highest extraction efficiency.

Influence of sample volume
The Cd 2+ solution volume is an important factor for achieving a high EF and maximum recoveries based on the VA-IL-DLLME procedure.
The sample volume effect was investigated using (5.0-50 mL) model solutions (Fig. 5). Higher than 25 mL, Cd 2+ ion recoveries were not quantitative. Hence, the Cd 2+ solution (25 mL) was chosen as the highest sample volume in all the subsequent studies. The PF has been described as the sample volume ratio to the final dilute volume of the IL phase (0.25 mL). Therefore, PF was attained at100.

Influence of vortex time
In the microextraction technique, the vortex has a significant effect on the dispersion of the IL phase into the aqueous phase and improves the extraction efficiency [29,30]. The impact of vortex time was investigated between 0.5 and 5.0 min at a constant rate 3000 rpm (Fig. 6). The results indicate that the extraction efficiency was increased up to 1.0 min. After this time, no significant improvement in analytical signals was obtained. As a result, 1.0 min was chosen as the optimum vortex time, which was adequate to fully dissolve the IL in the aqueous phase.

Influence of matrix ions
The possible effect of the commonly present matrix constituents on the preconcentration and determination of the Cd 2+ ions in different real samples was examined. The results obtained are shown in Table 1.
The tolerance limit is defined as the highest amount of interfering ions creating a relative error ≤ ± 5%. An advantage of this method was that the Cd 2+ ions recovered from the multi-element real samples were quantitative (>95%) at the level of tolerance, as provided in Table 1. No obvious interference effect of matrix ions was observed in the determination of Cd 2+ ions under the experimental conditions which confirm the applicability of the VA-IL-DLLME method for Cd 2+ determination in various real samples.

Analytical features of the proposed method
Using the optimized experimental conditions described above, a satisfactory linear relationship was obtained in the range of 1.0-300 µg/L with the following linear regression equation, A =7×10 -4 C+1.2×10 −3 with a correlation coefficient (r 2 =0.9997), where A is the absorbance and C is the Cd 2+ concentration (μg/L). The LOD and limit of quantification (LOQ) were calculated as 3 sb/m and 10 sb/m, respectively, where sb is the standard deviation often replicates measurements of blank samples and m is the slope of the calibration curve. The LOD and LOQ were 0.1 and 0.33 µg/L, respectively. The lower detection limit of the present VA-IL-DLLME method indicates high sensitivity and suggests its efficient application for the determination of very low concentrations of Cd 2+ in real samples.
The performance of the proposed VA-IL-DLLME procedure was assessed by calculating two parameters, including the EF and the consumptive index (CI). The EF defined as the ratio between the calibration graph slopes with and without preconcentration procedure (EF=30.43). The CI was determined using the expression CI = Vs/ EF, where Vs is the analyte solution volume [30] and CI is 0.82. The reliability and precision of the proposed UA-IL-DLPME system as the relative standard deviation (RSD%) were examined by applying five replicate determinations of 25, 150, and 250 µg/L of Cd 2+ , and RSD% of the recoveries was found to be 2.0-3.2%, which illustrate a good precision of the method ( Table 2).

Validation studies
In addition, the applicability of the developed VA-IL-DLLME method was checked and validated for the estimation of Cd 2+ concentrations in certified reference materials (SRM 1570A spinach leaves) through the standard addition method at three concentrations. The concordant resulting values of recoveries levels derived from the method developed were in good agreement with the CRMs certified values ( Table 3). The application of the proposed method to the real samples and CRMs shows that this method is highly accurate and reliable, is free of interference and validates the estimation of Cd 2+ ion at trace levels in real samples. The t crit values for 2° of freedom at the confidence level of 95% were statistically higher than the t exp . It indicates that the results showed no significant error (Table 3).
Another study for validation of the proposed method, recovery check for spiked samples was performed intraday and interday analysis. Five duplicate extractions and analysis of three concentrations of Cd 2+ solutions (75, 150, and 225 µg/L) were performed on the same day to assess the RSD% for the intraday precision. Furthermore, for analyzing the same Cd 2+ concentrations, the precision (as RSD%) for interday analysis was determined by extractions on 5 consecutive days. The RSD% ranged from 1.15-1.74% to 0.88-1.40% for intraday and interday precisions, respectively. The method's reliability (as recovery%) was reached in the range of 95.0-98.0% (Table 4).

Analytical applications to real samples
The potential application of the developed VA-IL-DLLME preconcentration method for the identify and separate of Cd 2+ ion in different real acid digested samples, including food (rice, tomato paste, white bread, onion, potato, spinach, lettuce, cabbage, and apple) samples and fruit juice (orange, apple, and grape) samples was tested.
The sample solutions were spiked with known quantities of Cd 2+ ion using the standard addition method to verify the reliability and accuracy of the developed process. Table 5 shows the percentage of the analyte recovered from the spiked real sample and the RSD%. The quantitative recoveries for the Cd 2+ analyte were great, in the range of 95.0-100% with RSD < 1.14%. Such results show that the approach is accurate and could therefore be used to separate, preconcentrate, and evaluate trace amounts of Cd 2+ in real food samples.

Comparison of the proposed method with other methods
The comparative data of different analytical characteristics showed that the obtained LOD, EF, and RSD of the proposed method are much better than other reported preconcentration methods for Cd 2+ determinations [1,3,22,[31][32][33][34][35][36][37] (Table 6). These results demonstrated that the proposed VA-IL-DLLME procedure could be successfully applied to the analysis of different real samples without any systematic error.