Two-Steps Cloud Point Extraction-Spectrophotomtric Method for Separation, Preconcentration and Determination of V (IV) and V (V) Ions in Real Samples Using Laboratory-Made Organic Reagents

The study was carried out in complete cooperation between all authors. Authors ZAAK and KJA designed and supervised the work study. Author ZAAK wrote the protocol, helped in analyzing the data statistically and wrote the final draft of the manuscript. Author KJA helped in the preparation of the new organic regents. Author ZTI carried out the most experimental works according to the cited plan and managed the literature searches. All authors have been read and approved the final ABSTRACT Aims: To establish a new analytical method for the extraction and pre-concentration of V (IV) and V (V) species in real samples by cloud point extraction (CPE) coupled with spectrophotometry using two newly laboratory-made chelating reagents. Study Design: All factors affecting the extraction and determination of V (IV) and V (V) ions using micelle-mediation Methodology: The approach is based on sequential separation of two vanadium species in the same solution. First, the complexation of vanadium (IV) with 2-[(Benzo imidazolyl) azo]-4-benzyl phenol (BIABP) at pH 3.0 and then extracted into micelle phase. Second, the vanadium (V) remaining in aqueous phase after the separation of (IV) is complexed with 2-[2-(5-Nitro thiazolyl) azo]-8 hydroxyquinoline (5-NTA8HQ) and H 2 O 2 in acidic medium to form a ternary complex (V (V) H 2 O 2 -NTA8HQ) which being re-extracted into micelle phase of Triton X-114. The extracted complexes in cloud point layer are dissolved in a minimum amount of ethanolic 0.1 M HNO 3 , then V (IV) and V (V) are determined spectrophotometrically at their respective absorption maxima. The proposed method was applied to the estimation of the two vanadium species in various real samples with satisfactory results where the method detection limit in these matrices was of 0.120 and 0.037 µg g -1 for V (IV) and V (V) respectively. Results: At established optimized conditions, a 159 and 99 fold enrichment factors and linear range of 10-100 and 1-70 ng mL -1 , leading the limits of detection of 1.78 and 0.75 ng mL -1 for V (IV) and V (V) ions respectively to be achieved in aqueous solution. The average percent recovery of 98.3±0.7 and 97.6±0.4 and a precision (RSD%, n=8) of 0.67% and 0.46%, at 40 and 30 ng mL -1 for V (IV) and V (V) are obtained. Conclusion: The described method is sensitive, easy to apply and interferences-free and in that way the determination of vanadium species in different samples was easily achieved. The results of the established method were compared statistically with ETA-AAS using t-paired test showing no significant difference at 95% confidence interval and the proposed method gave comparable analytical figures of merit compared with other sophisticated techniques.


INTRODUCTION
In our recently published paper concerning the speciation analysis of iron by using CPE-Spectrophotometry [1], we have made clear in detail the importance of challenges facing the analysts in the analysis of various oxidation states of metals and we focused on the difficulty of choosing the appropriate analytical method, especially when metal species present at low concentration level and in the complex matrices. However, we have been able to overcome some of the difficulties in the analysis of iron species, since the results were satisfactory and worthwhile, which encouraged the authors to engage in more complicated topic, namely the separation and determination of vanadium species by using the same above methodology. The main reasons for the choice of vanadium in this work are its analysis difficulty which lies in the possibility of redistribution of vanadium species, particularly when the environment of the sample is changing [2], its biological and environmental importance and the few papers published in chemical literatures related to the use of cloud point extraction as compared to more common elements such as Fe, Hg, As and Se. Vanadium exists in various oxidation states, but the most two common species occurring in environmental and biological systems are vanadium (IV) and vanadium (V) [3][4]. Recently, vanadium is deemed to be an essential element at trace level and plays a potential role in human health especially with diabetics of type I and II [5]. However, at a high concentration level, vanadium compounds could be highly toxic to humans and animals which will affect kidney and liver functions [6]. However, its toxicity depends on its oxidation state, for which V (V) being more toxic than V (IV) and V (IV) as vanadyl sulfate is 6-10 times less toxic than V (V) as vanadate [7][8]. In fact, there is no recommended intake levels of vanadium established by international bodies so far. However, based on negative impacts observed in animal studies, has been placed accepted high intake of a standard of vanadium is 1.8 mg per day [9]. Exposure of vanadium to humans occurs via different sources, including water, which is a good indicator of urban pollution levels. Most food such as milk, grains, cereals and vegetable oils is rich in vanadium, while fruits, meats, fish and batter are relatively poor sources of vanadium. In view of these facts, we find that the speciation analysis of vanadium is of extreme importance and represents one of the keys to the understanding of their possible harmful effects to biota and humans [10].
To the best of our knowledge, there is no paper dealing with the assessment vanadium in both forms using the cloud point extraction (CPE) methodology combined with spectrophotometry so far. In this work, an attempt was made to adopt two newly synthesised organic reagents prepared in our laboratory to embark on the determination of V(IV) and V(V) ions in some food and environmental samples by combined CPE-spectrophotometry, perhaps can we contribute to expand the horizons of applications of this methodology in analytical chemistry.

Apparatus
Two spectrophotometer systems were used in this study namely, a PG Instrument T80+ UV/Vis spectrometer (England) and UV-7804C (China) equipped with a 10-mm quartz cell, for recording of absorption spectra of the complexes formed and absorbance measurements respectively. A double-beam Atomic Absorption Spectrophotometer novAA-400 (Analytic Jena, Germany) equipped with graphite furnace and provided with Ultra fast background correction using a Deuterium lamp and vanadium hallow cathode lamp (operated at 5 mA) as the radiation source at the wavelength of 318.4 nm with 0.5 nm spectral band pass was used for determination of V species. The mass spectrometric measurement of the prepared ligand was carried out by using an Agilent 5975C inner MSD mass spectrometer (J and W Scientific Agilent Technologies, USA) at University of Tarbiat Modares, Tahran, Iran. The mass spectrum was obtained in electron-impact mode (EI) at 70 eV and a direct insertion probe (Acq method 10 W energy) at temperature 90-110ºC. The cloud point temperature of the surfactant was monitored via using a microprocessor-controlled water bath WB 710 model with temperature accuracy of ±0.3ºC at 37ºC (OPTIMA, Japan). The pH meter Philip PW model 9421 (Holland) equipped with combined electrode was employed for solution pH during the optimization and measurement steps.

Materials and Reagents
The chemicals used in this work including, a 2amino-benzimidazole and p-benzyl phenol (Riedel-deHaën, Seelze, Germany), sodium nitrite, hydrochloric acid, sodium hydroxide, acetic acid and sodium acetate (BDH, England), ethanol (GCC, England), vanadyl sulfate and ammonium metavanadate (Merck, Germany) and Triton X-114 (ACROS ORGANICS, New Jersey, USA) were used as received without any further purification. Doubly distilled and/or deionized water used throughout. The stock solutions (1000 μg mL

Synthesis and Characterization of Reagents
The synthesis and characterization of the first ligand namely 2-[2-(5-Nitro thiazolyl) azo]-8 hydroxyquinoline (5-NTA8HQ) used in this work was described in our previously published paper [1]. The second ligand of 2-[(Benzo imidazolyl) azo]-4-benzyl phenol (BIABP) was synthesized according to the procedure described elsewhere [30] with some modifications as showed in Fig. 1 ); IR(KBr) ν max /cm (BIABP) reagent solution for V (IV), 0.5 mL of acetate buffer solution at pH =3.0 and 0.2 mL of (10% v/v) Triton X-114 was kept in a thermostatic water bath at 70ºC for 10 min and two phases were separated by centrifugation for 5 min at 4000 rpm min. The viscosity of the surfactantrich phase was increased by cooling the system in an ice-bath for 20 min. The supernatant aqueous phase was carefully collected and left aside for the subsequent extraction of V (V). The surfactant-rich phase was dissolved with a 2 mL of 0.1 M nitric acid in ethanol and the concentration of V (IV) ions, was determined spectrophotometry at λ max of 625 nm. The abovementioned supernatant containing V (V) in an acidic medium (pH ≈2.5) was taken, then 0.5 mL (NTA8HQ) reagent and 0.2 mL of (10% v/v) Triton X-114 were added and the resultant solution were held for 10 min in a thermostatic bath at 70ºC and the two phases were separated by centrifugation for 5 min at 4000 rpm. On cooling in an ice bath, the surfactant rich phase became viscous and the supernatant aqueous phase was carefully removed with a pipette. The surfactant-rich phase was dissolved as with the above-mentioned in V (IV) extraction and the concentration of V (V) ions was determined spectrophotometry at λ max of 634 nm.

Water [20]
The water sample was collected and preserved by the addition of 2 mL of concentrated nitric acid in a polyethylene container that had been carefully cleaned with nitric acid. The samples were stored at 4ºC prior to the measurements. 2 mL of sample was taken and the concentration of V (IV) and V (V) was determined according to the general CPE procedure.

Soils
The samples were dried and grounded into fine powder using a glass mortar, then an accurately amount of 1.00 g of a powdered soil sample was transferred into a 25 mL platinum crucible and digestion procedure was carried out in accordance to the protocol of Molathegi [33]. The sample was first treated with 10.0 mL of HF and 2.0 mL of HClO 4 and evaporated till near dryness. Subsequently a 2 mL HF and 1 mL of HClO 4 were added and the mixture again evaporated to near dryness. Finally, HClO 4 (1 mL) was added and the sample was evaporated until white fumes appeared. The residue was then dissolved in 5 ml of 6 M HCl and diluted to 50.0 mL with de-ionized water. Extraction and determination of vanadium species were performed by the proposed method.

Rice
The rice sample solution was prepared according to the procedure adopted by Swetha el at. [34] by taking 1.00 g of dried rice at 110ºC and digested with 10 mL of 5M HNO 3 followed by addition of 5 mL of HClO 4 (70% w/w). The solution was evaporated to near dryness and the residue was dissolved in 10 mL of 0.1M HCl. Then heated to boiling, cooled and filtered. The filtrate was transferred into a 50 mL volumetric flask and diluted to the mark with deionized water. An appropriate amount of this solution (i.e. 2 mL) was taken and subjected to the CPE procedure to determine vanadium species spectrophotometrically.

Vegetables
Three vegetables samples (potato, spinach, carrot) were collected from local markets and subjected to the procedure described by Lokeshappa et al. [35] with little modification. Dried vegetables were first air-dried in an oven at 105ºC for three days under uncontaminated conditions. A 1.00 g of each grounded powder sample were transferred into glass beaker and digested with 10 mL of HNO 3 and 2.5 mL of HCl (4:1 v/v) on a hot plate at low temperature till complete digestion. After cooling at room temperature, the solution was made to 50 mL with deionized water in a 50 mL volumetric flask. An aliquot (2 mL) of the sample solution was taken, followed CPE procedure and the amount of the two species was determined spectrophometrically.

Absorption Spectra
The spectroscopic study was conducted by recording the absorption spectra of [V (IV) -BIABP] complex in the presence of surfactants versus a reagent blank. Fig. 3 shows the spectra of V Concerning the stoichiometric ratio of the ternary complex of type V (V)-H 2 O 2 -NTAHQ, the majority of reports have indicated that the molar ratio and continuous variation methods did not represent a true stoichiomtric value for this type of complexes. So the mathematical method described by He et al. [36] which it did not mention in detail in this study, because of the large number mathematical derivations, was conducted to determine stoichiometric ratio of V (V)/H 2 O 2 /NTAHQ system. This method was confirmed that the mole ratio of this type of complex is 1:1:1, from which the proposed chemical structure of the ternary complex can be deduced as shown in Fig. 6.

Factors Affecting CPE Procedure
The effect of the factors such as, pH, concentration of H 2 SO 4 , H 2 O 2 quantity, concentration of reagents, surfactant amount, temperature and incubation time were searched using the classical optimization strategy to obtain the optimum conditions which can achieve the best analytical figures of merit for the two species. All these experiments were performed for the solutions containing 60 ng mL -1 V (IV) and/or 30 ng mL -1 V (V). H 2 SO 4 . The results depicted in Fig. 8 showed that the absorbance signal reaches a maximum at 0.05 M H 2 SO 4. At higher concentrations of H 2 SO 4 , the absorbance decreases which most probably due to the reduction of V (V) ion to (IV) thus preventing the formation of ternary complex in micelle-mediated phase. Therefore, a 0.05 M of H 2 SO 4 was used for further experiments. The influence of H 2 O 2 concentration on the formation of V (V)-H 2 O 2 -NTA8HQ ternary complex was carried out by varying volume from 0.1 to 1 mL of H 2 O 2 (1% v/v) as shown in Fig. 9. It was observed that the analytical responses increase rapidly as the volume of H 2 O 2 increases and reach maximum up to 0.5 mL and it remained constant for higher added concentrations. Therefore, 0.5 mL of H 2 O 2 was selected as optimal. The effect of the (5-NTA8HQ) and (BIABP) concentration was conducted for the solutions containing 30 ng mL −1 V (V) and 60 ng mL −1 V (IV) and varying volume from 0.1 to 1 mL of 1 x 10 -2 M (5-NTA8HQ) and (BIABP). In both cases, V (V) or V (IV), the analytical responses increase rapidly as the volume of (5-NTA8HQ) or (BIABP) increases and reaches maximum up to 0.3 mL and 0.5 mL of 1.0 × 10−2 M of (NTA8HQ) and (BIABP) respectively and decrease thereafter with further increase in the chelating agents indicating that any excessive amount of chelating reagents was unnecessary (Fig. 10). Consequently, 0.3 mL of 1 x 10-2 M of (5-NTA8HQ) and 0.5 mL of 1 x 10-2 M of (BIABP) was chosen as optimum for V (V) and V (IV) respectively. Fig. 11 shows the impact of Triton X-114 amount on extractability of the two complexes within the surfactant volume range of 0.1 -0.4 mL of 10% (v/v) Triton X-114 at previously established optimum conditions. It can be seen that the absorbance for both ions increased by increasing the Triton X-114 concentration up to 0.2 of 10% (v/v) for V (V) and V (IV) and then suddenly decreased at higher amounts. Thus 0.2 mL of 10% (v/v) Triton X-114 was used as the optimum amount for V (V) and V (IV) for subsequent experiments. Fig. 12 shows the influence of the equilibrium temperature ranged from 30 to 80 at 10 min on extraction of the two complexes by CPE. It was shown that a maximum absorbance signal was achieved when the temperature at 70ºC for both species. Whilst Fig. 13 displays the effect of incubation time and found to be 10 min needed for complete extraction of both species in their complexes.

Calibration Graphs and Statistical Treatments
The calibration graphs for both species were constructed at the established optimized conditions using combined CPE-Spectrophotometry by taken a series of standard V (IV) and V (V) solutions ranging from 10-100 and 1-70 ng mL -1 respectively. The two calibration plots were subjected to the statistical evaluation which shown that a strong correlation exists between the calibration points (r = 0.9997 and 0.9999 for V (IV) and V (V) respectively) as shown in Table 1. This was supported by ANOVA analysis (Table not shown) giving that MS reg /MS error = 22870 for V (IV) and 106582 for V (V) for 1 and 8 dof and 1 and 7 dof, larger than critical value (F 1, 8 = 5.32 and F 1, 7 = 5.5 9 at 95% CI) and confirmed by the normal probability plots (Fig. 14, a and b) which revealed that an ideal linear trend indicative of normality of absorbance response being acceptable and statistically valid [37]. The calibration plots for both species have also been undergone the statistical treatments to extract new analytical figures of merit which are summarized in Table 1.

. Effect of temperature on the extraction of V (V)/V (IV) complexes by CPE
It can be seen from Table 1, high enrichment factors were achieved which reflected in enhancement of the sensitivity in term of molar absorptivity or Sandell sensitivity of the proposed method via using has new synthesised chelating agents. This obviously led to obtain very low detection limit in order of 2.53 and 0.72 ng mL -1 for V (IV) and V (V) ions in aqueous solutions respectively. These findings were much better than that obtained by other workers (   This leads us to be quite sure that the prepared ligands have had a major role in enhancing the sensitivity and lowering the detection limit of vanadium species by the proposed method, which gave an impression that there is no need to use much sophisticated instrumentation compared with CPE-Spectrophometry as described by our previous studies [42][43][44][45], concluding that this will easily motivate of using the proposed method in the detection of vanadium species in foods, drug formulations and environmental fields alike. The LOD's of the method were also calculated and found to be 0.120 and 0.037 µg g

Accuracy Evaluation
Due to the universal lack of certified reference materials that define exactly the quantity of vanadium species, the accuracy of the established method was carried out by assessing the recovery percentage by taking a binary mixture V (V)/V (IV) solution in concentration ratio range from 0.1-2. The results are summarized in Table 3.

Interferences Study
The selectivity of the suggested method was tested against the potential of some divers' metal ions which may affect the determination of the vanadium species in the selected samples. In this regard, the effect of 1000 fold concentration of each interfering ions on the estimation of 60 ng/mL of V (IV) and 40 ng/mL V (V) solutions were studied following the general CPE procedure. The results are summarized in Table 4.
It can be concluded that there are no appreciable influences for any of interfering ions on the responses of both vanadium species (Table 4). This might be attributed to inability of the interfering ions to form complexes with the two synthesized ligands at the working pH value of the proposed method.

Applications Study
According to the considerable analytical features that have been achieved in the proposed method, such as low detection limit, high recoveries and interference-free, the method was employed for the detection of both species in water, soil, rice and vegetables samples after the digestion procedures that described in experimental work and measured in triplicate. At the same time, the sample solutions were also determined by electrothermal atomic absorption spectrometric method (ETAAS) to test the significance of the proposed method. The findings are presented in Tables 5. The statistical computations using t-paired test for all tested sample reveals that the proposed method has no significant difference compared with ETAAS method at 95% confidence level.