Ionic Liquid Based Dispersive Liquid-Liquid ...

Send Orders for Reprints to [email protected] Current Analytical Chemistry, 2017, 13, 000-000

1

RESEARCH ARTICLE

Ionic Liquid Based Dispersive Liquid-Liquid Microextraction Combined with Magnetic-Based Dispersive Micro-Solid-Phase Extraction for Determination of Trace Cobalt in Water Samples by FAAS Hayati Filik* and Asiye Aslıhan Avan Istanbul University, Faculty of Engineering, Department of Chemistry, 34320 Avcılar, Istanbul, Turkey

ARTICLE HISTORY Received: November 16, 2016 Revised: February 01, 2017 Accepted: February 23, 2017 DOI: 10.2174/15734110136661703070934 52

Abstract: Background: Cobalt is one of the most important essential trace metals of human nutrition. Low doses of cobalt are needed for many humans and animals to stay healthy. Cobalamine (named vitamin B12) is a cobalt-containing essential vitamin. The direct detection of metal ions in various matrices by AAS may be difficult due to matrix interferences and extremely low levels of metal ions. Thus, a preconcentration and separation step is normally demanded. In this report, a two-step microextraction technique, combining room temperature ionic liquid based dispersive liquid-liquid microextraction (IL-DLLME) and dispersive magnetic solid-phase microextraction (MSPME) was prepared for the flame atomic absorption spectrometric determination (FAAS) of trace cobalt ions in water samples. Methods: In this study, a two-step microextraction technique based on a new combined approach of ILDLLME/M-SPME was discussed and the proposed method was applied to the rapid determination of trace cobalt ions in various water samples. In this study, room temperature IL—1-ethyl-2,3- dimethylimidazolium bis(trifluoromethyl sulfonyl)imide [EMIM][Tf2N] was employed as an extractant in the first extraction step (i.e. DLLME). This combined technique offers low limits of detection and high preconcentration factors resulting in high sensitivity. The reported method described an amazing and innovative approach of combining different microscope sample preparation methods to solve some analytical problems. Results: In the first microextraction step, room temperature ionic liquid (RTIL) was employed to extract cobalt-(5-bromo-2-pyridylazo)-5-diethylaminophenol (5-Br-PADAP) complexes from aqueous solution with ultrasound. In the second step, magnetic Fe3O4 NPs was added as an adsorbent and employed to collect the analytes in the organic solvent. After magnetic solid phase separation, the concentrated analyte complexes were eluted with 0.1 M HCl. As a consequence, the linear working range was 0.1-250 μg/L, and the detection limit of the method (LOD) was estimated to be 0.05 μg/L. The proposed two-step extraction procedure was employed by analysis of a certified reference and real water samples. Conclusion: In the current study, a simple two-step extraction method, namely ionic liquid DLLME combined with magnetic SPME was developed. The magnetic Citrate-Fe3O4 (Cit-Fe3O4) nanocomposite was synthesized and can be easily separated by a powerful magnet. In addition, this technique does not require clean-up steps and the magnetic sorbent material (i.e. after extraction, IL adsorbed onto the magnetic material) can often be easily regenerated and reused. This new extraction methodology offers various features such as sensitivity, cost-effective, easy to operate, short extraction time, low detection limit, and use less-toxic organic solvents. In fact, the preconcentration method was successfully applied for Co determination in water samples, with good accuracy and good reproducibility. The calibration graph of two-step microextraction method was linear in the range of 0.1-250 μg/L and the percent recovery of Co was in the range of 95-102 %. This is an applicable method for cobalt trace assay in various types of matrices in order to diminish its hazardous effects on ecosystem and environment.

Keywords: Cobalt, liquid-liquid microextraction, magnetic solid phase microextraction, atomic spectroscopy, water analysis. 1. INTRODUCTION Cobalt is one of the most important essential trace metals of human nutrition. Low doses of cobalt are needed for many *Address correspondence to this author at the Istanbul University, Faculty of Engineering, Department of Chemistry, 34320 Avcılar, Istanbul, Turkey; Tel: +90 212 473 70 70/17739; Fax; +90 212 473 71 80; E-mail: [email protected] 1573-4110/17 $58.00+.00

humans and animals to stay healthy. Cobalamine (named vitamin B12) is a cobalt-containing essential vitamin [1, 2]. Cobalamine stimulates the formation and production of red blood cells. It has also been employed as a therapy for anaemia. Cobalt poisoning can take place, particularly when people are exposed to high levels of inorganic cobalt which stimulates the growth of the thyroid gland. Greater quantities of cobalt salts are added to or naturally occur in food [2-4].

© 2017 Bentham Science Publishers

2 Current Analytical Chemistry, 2017, Vol. 13, No. 00

Cobalt salt is a chemical food additive that inhibits the formation of foam in many industrial process liquids [3]. The direct detection of metal ions in various matrices by AAS may be difficult due to matrix interferences and extremely low levels of metal ions. Thus, a preconcentration and separation step is normally demanded. The determination of trace metal ions has been performed by different modern instrumental techniques such as atomic absorption spectroscopy (AAS), inductively coupled plasma spectrometry (including ICP-AES, ICP-OES, and ICP-MS), voltammetry, ultraviolet-visible spectrophotometry (UV-vis), and neutron activation analysis (NAA), and highperformance liquid chromatography (HPLC) [5]. However, the quantification of trace metal ions in the aqueous samples usually requires a preconcentration step before detection due to the low levels of the elements in complex matrices. A current course in analytical chemistry is minimized of preconcentration systems with the aim of minimizing reagent consumption and waste generation. The detection limit of the target analyte can be substantially improved if the final sample volume is diminished. Recently, the enrichment of cobalt ions has been performed by different microextraction techniques such as cloud point extraction (CPE) [6-9], solid phase microextraction (SPME) [10], solidified floating organic drop microextraction (SFODME) [11-13], single drop microextraction [14], cold-induced aggregation microextraction (CIAME) [15], dispersive liquid-liquid microextraction [16-23], and deep eutectic solvent based ultrasonic assisted liquid phase microextraction [24]. The substantial benefits of microextraction techniques are rapidity, low price, high recovery, high enrichment factor and short extraction time. Previously, DLLME technique has been utilized to the enrichment of several families of inorganic and organic species prior to analysis with modern instrumental methods [25-29]. However, DLLME has drawbacks that contain the loss of organic solvent in the extraction step, inconvenience in recovering the organic phase, and relatively low recoveries of the target analytes [30]. Moreover, organic layer separation requires extra instrumentation (i.e., centrifugation, refrigeration, and filtration) and this tedious step leads to relatively low precision and makes preconcentration difficult to achieve in a routine assay. To eliminate these problems, a two-step technique based on DLLME and M-SPME has emerged as an effective and simple approach [30]. M-SPME is a solid phase extraction technique that has obtained considerable attention in the analysis of target analytes in real samples. M-SPME is based on the extraction of the different compound from the sample using solids with magnetic properties. This separation technique can be envisioned as a magnetic separation commonly employed to separate magnetic phases from nonmagnetic phases [31]. In addition, the M-SPME does not need additional steps such as centrifugation, precipitation, or filtration of the sample. Briefly, both methods have some advantages and disadvantages. A twostep DLLME and M-SPME procedure were combined for the separation and preconcentration of polycyclic aromatic hydrocarbons (PAHs) [30]. Recently, Wang et al. [32] introduced a two-step microextraction method for simultaneous preconcentration of various hard metals such as Cd, Ni, and Cu based on the complex formation with the diethyldithiocarbamate. We developed a two-step extraction method for

Filik and Avan

measuring of trace cobalt concentrations in environmental water samples. The new microextraction approach was generally performed in two separate steps. In the first step, an organic solvent (ionic liquid) is employed in DLLME for the isolation and preconcentration of target analytes from aqueous samples. In the second step, unmodified or modified Fe3O4 MNPs have added to recovery the organic solvent that included the analytes in the presence of a magnetic field. Yet, until now, IL-DLLME integrated with M-SPME has not even been described for the detection of trace cobalt ions by FAAS. In this study, a two-step microextraction technique based on a new combined approach of IL-DLLME/M-SPME was discussed and the proposed method was applied to the rapid determination of trace cobalt ions in various water samples. Ionic liquids are made entirely of ions and they possess charming features, such as good electrolytic conductivity, tunable miscibility and viscosity, and low volatility, which make ILs useful and matchless for many applications in the chemical assay. RTILs have been widely carried out in most subdisciplines of analytical chemistry, such as sample preparation, separations, and chemical analysis [33-34]. In this study, room temperature IL-1-ethyl-2,3-dimethyl imidazolium bis(trifluoromethyl sulfonyl)imide [EMIM][Tf2N] was employed as an extractant in the first extraction step (i.e. DLLME). This method presents a linear range and a detection limit that is comparable to, or better than other strategies developed for Co-determination in food, biological and environmental samples [6-10, 15, 17-22, 24] (Table 1). The results indicate that the proposed method is a simple, fast, interference-free, selective and environment-friendly analytical approach for trace Co-determination in aqueous samples. Besides, the reported method described ensures an amazing and innovative approach of combining different microscale sample preparation methods to solve some analytical problems. 2. EXPERIMENTAL 2.1. Apparatus The determination of Co(II) was employed using the Varian Spectra AA-400 AAS (Agilent, Santa Clara, CA, USA) equipped with a deuterium background correction and an air-acetylene burner. Plastic centrifuge tubes (10 mL) were used for phase separation steps. The analyte solution pH values were controlled by a pH-meter (Hanna HI 221, Midland, ON, Canada). The powder X-ray diffraction (XRD) patterns were recorded with a Rigaku (D/MAX-2200/PC) diffractometer equipped with source Cu Ka radiation. The FEI Quanta 450 FEG (FEI Company, USA) scanning electron microscope (SEM) is used for high-resolution imaging. Ultrasonic bath (operating frequency, 50 ± 2kHz) was obtained from Isolab (Istanbul/Turkey). 2.2. Reagent and Solutions All chemical products were of analytical purity (purchased from Merck or Sigma-Aldrich) and distilled water was applied throughout each experiment. The Co (II) (110-3 M) stock solution was prepared by dissolving of the convenient amount of CoCl2·6H2O (Merck) in 0.01 M HCl. After

Ionic Liquid Based Dispersive Liquid-Liquid Microextraction Combined

Table 1.

Current Analytical Chemistry, 2017, Vol. 13, No. 00 3

Characteristic performance data obtained by using the proposed method and other preconcentration techniques for determination of cobalt in real samples. Linear

Detection

Range

Limit

(g/L)

(g/L)

20-200

7.5

[6]

CPE / TS-FF-AAS

2.1-100

2.1

[7]

CPE /FAAS

0.25-5.0

1.0

[8]

CPE/FAAS

0.9-100

0.9

[9]

SPE/FAAS

18-900

0.8

[10]

CIAME/FAAS

0-50

0.8

[15]

DLLME/Specrophotometry

2-50

0.5

[17]

DLLME/ICP-OES

NR

0.2

[18]

DLLME/FAAS

20-500

13

[19]

Microextraction Technique

CPE/Specrophotometry

Reference

DLLME/FAAS

3-100

0.9

[20]

DLLME/FAAS

10-150

2.42

[21]

IL-DLLME/FAAS

2-166

0.7

[22]

DU-LPME/FAAS

0.25-2.0

1.1

[24]

DLLME/MSPME-FAAS

0.1-250

0.05

This work

that, the stock solution was complexometrically standardized with EDTA (Merck). The calibration solutions were prepared by diluting the standardized stock solution. A solution of 5-Br-PADAP (1.010-3 M) was fixed in ethanol. [EMIM][Tf2N] (Sigma-Aldrich, 98%) was used as an extractant. A 0.01 M Na2EDTA solution was made in distilled water. Lower concentrations of the reagent were prepared by serial dilution. Buffer solutions were prepared by dissolving of the convenient amount of CH3-COOH/CH3-COONa (pH=3.6-6.5), and NaOH/KH2PO4 (pH= 6.5-12.0) (SigmaAldrich, 98%). Ammonium iron (II) sulphate, ammonium iron (III) sulphate, sodium citrate, and NaOH were purchased from Merck (Darmstadt, Germany). 2.3. Preparation of Fe3O4 Magnetic Nanoparticles In this work, synthesis of Fe3O4 magnetite nanoparticles was actualized as reported in the literature [35]. The Fe3O4 NPs were prepared based on the co-precipitation of Fe2+ and Fe3+ with a molar ratio of 2:1 (2.5 g (NH4)2Fe(SO4)2·6H2O and 5 g FeNH4(SO4)2 (Fe2++2Fe3++ 8OH- Fe3O4 + 4H2O) [35-38]. Typically, Fe(II)/Fe(III) (molar ratio of 2:1) mixture and 1.0 mL of concentrated hydrochloric acid were dissolved in 10 mL of distilled water degassed with nitrogen. Thereafter, the mixture of (Fe(II)/Fe(III) solution was added dropwise into 10 mL of 3 M NaOH solution under vigorous stirring in a nitrogen atmosphere. Subsequently, the solution stirred for 30 min at 70°C under the N2 protection. Then, 5 mL of an aqueous solution of sodium citrate (2.5 g) was added to the above reaction mixture and heated for 15 min at 70°C under the N2 protection [35-38]. After that, the precipitated black magnetic Cit-Fe3O4NPs were isolated by placing a strong magnet and washed repeatedly with distilled water and ethanol until pH neutral. Last, the obtained magnetic CitFe3O4 NPs were dried for 4 h at 50°C.

2.4. Procedure The analysis sample or the standard solution including Co(II) in the range of 0.1-120 μg/L and 5-Br-PADAP (2.010-6 M) was discharged in a 10.0 mL plastic centrifuge tube and the pH of the analyte solution was calibrated to 5 by acetate buffer solution. After 2 min, the reddish Co-5-BrPADAP complex occurred. The standard EDTA solution (110-3 M) should be added 5 min after the chelate formation (Co-5-Br-PADAP; metal/ligand ratios equal to 1:2) and the volume of the analyte was regulated to 10.0 mL with distilled water. The order of addition of the reagents can directly affect the sensitivity. After rapidly injecting 200 L of acetone extract (as disperser solvent) containing 100 L of [EMIM] [Tf2N] (as an extraction solvent) into the water sample, immediately, a cloudy solution was formed in the test tube. The mixture was kindly sonicated at room temperature for 1 min, and a steady cloudy solution was obtained. After 1 min, 30 mg of Fe3O4 NPs was rapidly added and the mixture was sonicated for 2 min at 50 kHz. During extraction, the ionic liquid was successfully immobilized on the support and in this case, the IL modified Fe3O4 also can be easily separated. After this extraction step, the IL/Fe3O4 NPs phase (probably IL coated Fe3O4) settled at the bottom of the tube. Furthermore, a strong magnet was placed at the bottom of the tube to collect the IL/Cit-Fe3O4 NPs, and the supernatant phase was readily decanted. After separation of the IL/Fe3O4 NPs, 0.5 mL of 1.0 M HCl (or 0.5 mL acetone) was transferred into the centrifuge tube in order to desorb the Co5-Br-PADAP complex by sonication for 3 min. After the desorption process, the IL modified Cit-Fe3O4 NPs were separated from the aqueous solution by the magnetic field and the eluted phase (i.e. HCl or acetone) was injected into FAAS for the determination of Co(II). The above procedure was implemented in the blank solution.


3. RESULTS AND DISCUSSION 3.1. Characterization SEM image of Fe3O4 is shown in Fig. (1). The bare Fe3O4 particles show a high degree of agglomeration due to dipoledipole interaction. Magnetic Fe3O4 NPs are spherical in form.

Filik and Avan

from aqueous solution was examined in the range of 3.6 to 10.0. The results illustrated that the recovery of Co is practical remains constant in the tested pH scale. Fig. (3) illustrates the influence of pH on the recovery of Co (II). According to the reached results, the optimum pH values were in the 3.6-10 range, inconsistency with the optimal complex formation range [39-41]. At lower pH values than 3.6, the formation of the Co-5-Br-PADAP chelate was not complete, so the two-step extraction yield was lower. This two-step extraction approach is workable over a wide pH range. As an outcome, pH 5.0 was employed as the working pH value further twostep extractions.

Fig. (1). SEM image of bare Fe3O4 NPs synthesized using alkaline co-precipitation method.

The XRD patterns of bare Fe3O4 NPs prepared by the coprecipitation method and the Fe3O4 retrieval citrate composites are shown in Fig. (2a and b). In the 2 region of 20-700, six characteristic peaks marked by their indices (220), (311), (400), (422), (511), and (440) were observed for bare Fe3O4 NPs and the Fe3O4 retrieval citrate composites. Because of the encapsulation by citrate on the magnetic nanoparticles, the peak intensities decreased (Fig. 2b) slightly from Fe3O4 NPs to the Fe3O4 retrieval citrate composites. The results are in good agreement with literature data [37, 38].

Fig. (3). Effect of pH on the extraction recovery of cobalt. Conditions: Sample volume 10 mL, Cobalt 2.95 g/mL, 5-Br-PADAP 5.010-4 mol/L, [EMIM][Tf2N] 100 L, Fe3O4 30 mg.

3.3. Effect of the Amount of Fe3O4 The synthesized citrate coated Fe3O4 NPs were used due to their perfect adsorption capacity, and masses between 10 and 100 mg were carefully tested. The cobalt recoveries were increased from 10 to 25 mg and remained remarkably constant at higher Fe3O4 masses. The extraction recoveries increased with increasing Fe3O4 NPs; further increasing the amount of the Fe3O4 NPs showed no significant improvement for the recoveries of cobalt ions. As a result, an amount of 30 mg of magnetic Fe3O4 NPs was chosen as optimum for consequent experiments. 3.4. Effect of 5-Br-PADAP Concentration

Fig. (2). XRD patterns obtained from bare Fe3O4 NPs synthesized using coprecipitation method. (a) Fe3O4 NPs, (b) citrate modified Fe3O4 NPs.

3.2. Effect of pH The analyte solution pH is a critical factor in IL-DLLME of Co using 5-Br-PADAP reagent because pH value is directly concerned with the creation of metal-ligand species. The influence of the pH on IL-DLLME of Co (2.95 μg/mL)

The extraction efficiency is directly dependent on the 5Br-PADAP concentration that is required to convert Co into an extractable form. The influence of the ligand concentration (i.e. 5-Br-PADAP) on the recovery was tested within the range 1.010-4-1.010-3 M. The maximum recovery was accomplished at a concentration of 510-4 M of the 5-BrPADAP and after that, the recovery stays almost constant. Thus, a concentration of 5.010-4 M 5-Br-PADAP was employed in the following experiments. However, the excessive complexing agent could be coextracted into the ionic liquid phase, thus decreased the extraction recovery of target analyte [41]. Fig. (4) shows the variation of recovery as a function of the 5-Br-PADAP concentration when 10 mL of solution including the specific amount of Co (2.95 μg/mL). In addition, the Co-5-Br-PADAP complex does not break down in the presence of EDTA and the colour of the chelate is remarkably stable between 5 min and 24 h in the RTIL phase.


Current Analytical Chemistry, 2017, Vol. 13, No. 00 5

[42, 43]. Based on these findings, acetone was chosen as a disperser solvent. To study the impact of acetone volume on the separation efficiency, several volumes of acetone in the range of 50-300 L were tested. During the investigation, the volume of [EMIM][Tf2N] was fixed at 100 L. According to the results, the recovery performance, increased with the increase of the disperser solvent volume when it is less than 200 L. By using more than 200 L acetone, the solubility of RTILs in water solution increases and the recovery performance decreased systematically. Therefore, 200.0 L of acetone was chosen as the optimum disperser solvent. 3.7. Sonication Time Fig. (4). Effect of 5-Br-PADAP concentration on the extraction recovery of cobalt. Conditions: Sample volume 10 mL, Cobalt 2.95 g/mL, [EMIM][Tf2N] 100 L, Fe3O4 30 mg, pH = 5.0.

3.5. Selection of Extraction Solvent and its Volume The choosing of a suitable extraction solvent is a considerable factor in DLLME process. In classical DLLME, mainly halogenated hydrocarbons (i.e. carbon tetrachloride, dichloromethane, chloroform, and chlorobenzene) are most suitable because they can be easily deported from the water matrix by centrifugation after extraction. Briefly, a highdensity solvent (i.e. extractant) is favourable extractants in classical DLLME. But an analyst at the highest risk of longterm exposure to these four volatile organic compounds. Recently, RTILs have been widely used as extraction solvents in DLLME. Conversely, RTIL is a suitable and lowtoxic organic extractant that has been extensively used in many applications. Thus, the impact of RTIL volume on the two-step extraction yield was performed with different volumes of RTIL in the range of 50-150 μL. As shown in Fig. (5), the best recovery was achieved when 100 μL of RTIL was employed. The recovery decreased with increasing RTIL volume due to the increasing volume of the RTIL phase. Under optimal conditions, the final phase volume was measured approximately 48±3 μL (n=5). As a result, 100 L of [EMIM][Tf2N] (as an extraction solvent) was employed in subsequent experiments.

Sonication was needed to deport the organic phase from the aqueous phase in a short time. Under ultrasonic energy, the rate of mass transport can be clearly raised. On the other hand, ultrasonic energy has a strong mechanical effect on the raw material surfaces, which can destroy the material surface and minimize the particle size [44]. In this setting, the effect of sonication time on IL-DLLME was tested. The sonication time was investigated at the time between the injection of the acetone containing RTIL and the start of sonication. Therefore, the influence of sonication time on the recovery of cobalt was explored in the range: 0-10 min. Complete twophase separation was accomplished within 1 min at 50 kHz. Thus, a sonication time of 1 min was picked as the optimal, since complete two-phase separation occurred in selected time period. No perceptible improvements were observed for longer sonication times. The consequences show that the microextraction process quickly reaches equilibrium, at an IL-DLLME time approximately 1 min. After the sonication process, 30 mg citrate coated Fe3O4 NPs quickly added to the extraction tube. For the M-SPME, the effect of ultrasonication time was also explored in the range of 0-5 min under fixed extraction circumstances. The percent recovery of the analyte increased with increasing sonication time up to 2 min, remaining almost stable at longer times. Consequently, 2 min of ultrasonication time was chosen for further experiments. 3.8. Desorption Studies

Fig. (5). Effect of [EMIM][Tf2N] volume on the extraction recovery of cobalt. Conditions: Sample volume 10 mL, Cobalt 2.95 g/mL, 5-Br-PADAP 5.010-4 mol/L, Fe3O4 30 mg, pH = 5.0.

In addition, the desorption time of cobalt was carefully examined. The desorption recovery is highly dependent on the type of desorption solvent. To find the best solvent, methanol, acetonitril, acetone and acetone/HCl (1 M) with different volume were tested and the highest recovery was obtained using acetone:1 M HCl (1:1). The results indicated that the recovery of cobalt ions increased with the increase in elution time, up to ca. 2 min and thereafter remained the same. As a result, the enough elution time of 2 min selected to be the most desirable. The results showed that acetone is necessary for the quantitative recoveries of analyte elements. Acetone is helped to break interaction between analyte chelates and adsorbent. All extractions and desorption tests were made at room temperature.

3.6. Effect of Volume of Disperser Solvent

3.9. Reusability of Magnetic Cit-Fe3O4 NPs

Acetone can be usually used to decrease the RTIL viscosity and the RTIL consumption in solvent extraction and it behaves as an almost universal organic solvent for RTILs

As adsorbent reusability is very important from both economic and environmental points of view. Consequently, the reusability of magnetic Fe3O4 NPs was tested in several sub-


Filik and Avan

sequent runs. After use, the magnetic Fe3O4 NPs should be carefully separated and washed three times with smaller portions of methyl alcohol and acetone to remove sorbed RTIL. The results indicate that the Cit-Fe3O4 NPs adsorbent could be reused for more than 5 sequential runs without any considerable decrease in adsorption capacity. The high reusability of the magnetic Cit-Fe3O4 NPs adsorbent can be explained by its high stability and broad surface area owing to an extremely high porosity. The above experiments indicated that the CitFe3O4 NPs could be regenerated for practical use. 3.10. Effect of Ionic Strength Generally, the adding of neutral salt in classical liquidliquid extraction using halogenated hydrocarbonsin creases the extraction efficiency due to salting out effect. A number of neutral salts are an important factor determining ionic strength. Different neutral salts may give rise to different ionic strengths. The outcomes demonstrated that the differences between the salting out effects of NaCl, KCl, NaNO3 , and KNO3 are not substantial. Therefore, NaCl was employed in further experiments. The effect of neutral salt was researched over a NaCl concentration range of 0-5% (w/v), while other microextraction variables remained stable during the investigation. In the range from 0 to 3%, the recovery of cobalt was found to be quite constant, while it decreased at concentrations higher than 3% (w/v) NaCl due to solubilization of the [EMIM][Tf2N] into the aqueous phase. The outcomes indicated that the neutral salt addition had no considerable impact on the extraction method. Therefore, entirely the subsequent assays were made without the addition of neutral salt. 3.11. Analytical Figures of Merit In order to test the feasibility of the suggested method for the quantitative assay of Co, the relationship between the absorbance and the concentration of Co was studied using FAAS. The calibration graph was linear between 0.1 and 250 μg/L and obeyed the equation A=0.0055 C (Co) + 0.0139 (R2 = 0.9985), where A is absorbance and C is cobalt concentration in μg L1. The LOD of the method was calculated to be 0.05 μg/L (i.e. identified as 3Sb/m). The relative standard deviation for a Co concentration of 50 μg/L was 4.4 % (n = 5). As a result, the DLLME/MSPME has a comparatively low detection limit, low consumption of an organic solvent, wider linear dynamic range and short extraction time [6-10, 15, 17-22, 24]. Table 2.

3.12. Interferences The influences of prospective interfering species were carefully investigated. A solution containing a known amount of Co (2.95 g/mL) and varying amounts of diverse ions was prepared, and the influence of some selected cations and anions on the two-step extraction system was examined at a fixed Co concentration under optimum twostep extraction conditions. The presence of most other metal ions does not interfere with the cobalt determination. Only Fe(II) caused serious interference, but it can be oxidized to Fe(III) by H2O2 and Fe (III) ions did not interfere with the Co(II) detection, because of the Fe(III) -5-Br-PADAP chelate is decayed totally in the presence of EDTA as a masking agent [38-40]. Masking agents (0.01 M) like citrate, EDTA, tartrate, and sodium fluoride do not interfere in the determination of Co. These two elements (i.e. Co and Ni) are always present together in real and synthetic samples. On the other hand, the coloured Ni-5-Br-PADAP complex was decomposed by heating the sample at 500C for 5 min after the addition of EDTA. According to the works of literature, at pH 4.75, the Ni-5-Br-PADAP complex was decomposed in 180 min, but at pH 4.0 only 90 min needed at room temperature [41]. Briefly, the proposed assay procedure is really sensitive. 3.13. Applications In order to show the feasibility of the two-step extraction approach, which was implemented to the determination of the Co content in the TMDA-51.3 certified reference water sample (the National Water Research Institute, Ontario/Canada). The content of Co obtained by the proposed method was 73.5 ± 3.0 μg/L (n=3), which agrees with the claimed value 71.5 ± 4.0 μg/L. This result demonstrates the applicability of the two-step extraction approach for Codetermination in aqueous samples. In order to confirm the method, which was implemented to real water samples. Mineral water and bottled water samples were obtained from a local market (Istanbul, Turkey). Water samples were normally transported to the laboratory and stored overnight in a refrigerator (< 40C) for chemical analysis the following day. Tap water sample was collected directly from the laboratory at the Department of Chemistry (Istanbul University/Turkey) and analyzed immediately after collection. The unspiked and spiked real samples were carefully analyzed, respectively, with the reached results shown in Table 2. The percentage of recovery of the spiked sample

Detection of Co (II) in tap and mineral, bottled water samples and relative recoveries of spiked samples. Sample Tap water

Mineral water

Bottled water

Added (g/L)

Found (g/L)

Recovery (%)

20.0 50.0 20.0 50.0 20.0 50.0

n.d. 19.4±1.5 49.1±2.3 n.d 20.3±1.6 51.1±2.2 n.d. 19.7±0.9 49.8±1.7

97 98 102 102 99 100


is in the range between 97 and 102 %, which clearly demonstrated the applicability and reliability of the new method. The results show that the suggested method is suitable for the quantification of Co in various aqueous samples.

Current Analytical Chemistry, 2017, Vol. 13, No. 00 7 [7]

[8]

CONCLUSION In the current study, a simple two-step extraction method, namely ionic liquid DLLME combined with magnetic SPME was developed. The magnetic Cit-Fe3O4 nanocomposite was synthesized and can be easily separated by a powerful magnet. In addition, this technique does not require clean-up steps and the magnetic sorbent material (i.e. after extraction, IL adsorbed onto the magnetic material) can often be easily regenerated and reused. This new extraction methodology offers various features such as sensitivity, cost-effective, easy to operate, short extraction time, low detection limit, and use less-toxic organic solvents. In fact, the preconcentration method was successfully applied for Co-determination in water samples, with good accuracy and good reproducibility. The calibration graph of two-step microextraction strategy was linear in the range of 0.1-250 μg/L and the percent recovery of Co was in the range of 95-102 %. This is an applicable method for trace cobalt assay in various types of matrices in order to diminish its hazardous effects on ecosystem and environment.

[9] [10]

[11]

[12]

[13]

[14]

ETHICAL APPROVAL This article does not contain any studies with human participants or animals performed by any of the authors. INFORMED CONSENT

[15]

[16]

Human and/or animal were not used in this work. CONFLICT OF INTEREST The authors confirm that this article content has no conflict of interest.

[17]

[18]

ACKNOWLEDGEMENTS We gratefully acknowledge Istanbul University Scientific Research Fund for financial support.

[19]

REFERENCES [1] [2]

[3] [4] [5]

[6]

Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological Profile for Cobalt. Public Health Service, U.S. Department of Health and Human Services, Atlanta, GA. 1992. California Environmental Protection Agency (CalEPA). Technical Support Document for the Determination of Noncancer Chronic Reference Exposure Levels. Draft for Public Comment. Office of Environmental Health Hazard Assessment, Berkeley, CA. 1997. Elinder, C.G.; Friberg, L. Cobalt. In: Friberg, L.; Nordberg, G.F.; Vouk, V.B., eds.; Handbook on the toxicology of metals, vol. II: Specific Metals, Elsevier Science: Amsterdam, 1986, 2, 211-232. Yamagata, N.; Murata, S.; Torıı T. The cobalt content of human body. J. Radiat Res., 1962, 3(1), 4-8. El-Shahawi, M.S.; Al-Saidi, H.M. Dispersive liquid-liquid microextraction for chemical speciation and determination of ultra-trace concentrations of metal ions. TrAC, Trends Anal. Chem., 2013, 44, 12-24. Safavi, A.; Abdollahi, H.; Hormozi Nezhad, M.R.; Kamali, R. Cloud point extraction, preconcentration and simultaneous spectrophotometric determination of nickel and cobalt in water samples. Spectrochim. Acta A, 2004, 60(12), 2897-2901.

[20] [21]

[22]

[23]

[24]

Donati, G.L.; Nascentes, C.C.; Nogueira, A.R.A.; Arruda, M.A.Z.; Nóbrega, J.A. Acid extraction and cloud point preconcentration as sample preparation strategies for cobalt determination in biological materials by thermospray flame furnace atomic absorption spectrometry. Microchem. J., 2006, 82(2), 189-195. Citak, D.; Tuzen, M. A novel preconcentration procedure using cloud point extraction for determination of lead, cobalt and copper in water and food samples using flame atomic absorption spectrometry. Food Chemical. Toxicol., 2010, 48(5), 1399-1404. Lemos, V.A.; da Franca, R.S.; Moreira, B.O. Cloud point extraction for Co and Ni determination in water samples by flame atomic absorption spectrometry. Sep. Purif. Technol. 2007, 54(3), 349-354. Ghaedi, M.; Ahmadi, F.; Soylak, M. Preconcentration and separation of nickel, copper and cobalt using solid phase extraction and their determination in some real samples. J. Hazard. Mater., 2007, 147(1-2), 226-231. Bidabadi, M.S.; Dadfarnia, S.; Shabani, A.M.H. Solidified floating organic drop microextraction (SFODME) for simultaneous separation/preconcentration and determination of cobalt and nickel by graphite furnace atomic absorption spectrometry (GFAAS). J. Hazard. Mater., 2009, 166(1), 291-296. Lan, J.; Zhao, Z. Determination of cobalt in water samples using solidified floating organic drop microextraction coupled with graphite furnace atomic absorption spectrometry. Chem. Spec. Bioavailab., 2012, 24(2), 124-128. Amirkavei, M.; Dadfarnia, S.; Shabani, A.M.H. Dispersive liquidliquid microextraction based on solidification of floating organic drop for simultaneous separation/preconcentration of nickel, cobalt and copper prior to determination by electrothermal atomic absorption spectrometry. Quím. Nova, 2013, 36(1), 63-68. Amjadi, M.; Manzoori, J.L.; Abulhassani, J. Ionic liquid-based, single-drop microextraction for preconcentration of cobalt before its determination by electrothermal atomic absorption spectrometry. J. AOAC Int., 2010, 93(3), 985-991. Baghdadi, M.; Shemirani, F.; Zhad, H.R.L.Z. Determination of cobalt in high-salinity reverse osmosis concentrates using flame atomic absorption spectrometry after cold-induced aggregation microextraction. Anal. Methods, 2016, 8, 1908-1913. Jiang, H.; Qin, Y.; Hu, B. Dispersive liquid phase microextraction (DLPME) combined with graphite furnace atomic absorption spectrometry (GFAAS) for determination of trace Co and Ni in environmental water and rice samples. Talanta, 2008, 74(5), 11601165. Gharehbaghi, M.; Shemirani, F.; Baghdadi, M. Dispersive liquidliquid microextraction and spectrophotometric determination of cobalt in water samples. Int. J. Environ. Anal. Chem. 2008, 88(7), 513-523. Dos Silva, E. S.; Correia, L.O.; Dos Santos, L.O.; Dos Santos Vieira, E.V.; Lemo, V.A. Dispersive liquid-liquid microextraction for simultaneous determination of cadmium, cobalt, lead and nickel in water samples by inductively coupled plasma optical emission spectrometry. Microchim. Acta, 2012, 178(3), 269-275. Ojeda, C.B.; Sánchez Rojas, F. Evaluation of dispersive liquidliquid microextraction for the determination of cobalt and cadmium by flame atomic absorption spectrometry: application in water and food samples. Sample Prep., 2014, 2(1), 13-20. Baliza, P.X.; Teixeira, L.S.G.; Lemos, V.A. A procedure for determination of cobalt in water samples after dispersive liquid-liquid microextraction. Microchem. J., 2009, 93(2), 220-224. Sorouraddin, M.-H.; Khoshmaram, L. Combination of Dispersive Liquid-Liquid microextraction with flame atomic absorption for determination of trace Ni and Co in water samples and vitamin B12. J. Chin. Chem. Soc., 2010, 57(6), 1346-1352. Abdolmohammad-Zadeh, H.; Ebrahimzadeh, E. Determination of cobalt in water samples by atomic absorption spectrometry after pre-concentration with a simple ionic liquid-based dispersive liquid-liquid micro-extraction methodology. Cent. Eur. J. Chem., 2010, 8(3), 617-625. Berton, P.; Wuilloud, R.G. Highly selective ionic liquid-based microextraction method for sensitive trace cobalt determination in environmental and biological samples. Anal. Chim. Acta, 2010, 662(2), 155-162. Arain, M. B.; Yilmaz, E.; Soylak, M. Deep eutectic solvent based ultrasonic assisted liquid phase microextraction for the FAAS determination of cobalt. J. Mol. Liq., 2016, 224, 538-543.

8 Current Analytical Chemistry, 2017, Vol. 13, No. 00 [25]

[26]

[27] [28]

[29]

[30]

[31] [32]

[33] [34]

[35]

Anthemidis, A.N.; Ioannou, G.K.-I. Recent developments in homogeneous and dispersive liquid-liquid extraction for inorganic elements determination. A review. Talanta, 2009, 80(2), 413-421. Herrera-Herrera, A.V.; Asensio-Ramos, M.; Hernández-Borges, J.; Rodríguez-Delgado, M. Dispersive liquid-liquid microextraction for determination of organic analytes. TrAC, Trends Anal. Chem., 2010, 29(7), 728-751. Ojeda, C.B.; Rojas, F.S. Separation and preconcentration by dispersive liquid-liquid microextraction procedure: A Review. Chromatographia, 2009, 69(11), 1149-1159. Rezaee, M.; Assadi, Y.; Millani, M.R.; Aghaee, E.; Ahmadi, F.; Berijani, S. Determination of organic compounds in water using dispersive liquid-liquid microextraction. J. Chromatogr. A, 2006, 1116(1-2), 1-9. Zang, X.-H.; Wu, H.; Zhang, M.-Y.; Xi, G.-H.; Wang, Z. Developments of dispersive liquid-liquid microextraction technique. Chinese. J. Anal. Chem., 2009, 37(2), 161-168. Shi, Z.-G.; Lee, H. K. Dispersive liquid-liquid microextraction coupled with dispersive μ-solid-phase extraction for the fast determination of polycyclic aromatic hydrocarbons in environmental water samples. Anal. Chem., 2010, 82(4), 1540-1545. Ibarra, I. S.; Rodriguez, J. A.; Galán-Vidal, C. A.; Cepeda, A.; Miranda, J. M. Magnetic solid phase extraction applied to food analysis. J. Chem(Hindawi). 2015, 2015, 1-13. Wang, X.; Chen, J.; Zhou, Y.; Liu, X.; Yao, H.; Ahmad, F. Dispersive liquid-liquid microextraction and micro-solid phase extraction for the rapid determination of metals in food and environmental waters. Anal. Lett., 2015, 48, 1787-1801. Sun, P. Armstrong, D. W. Ionic liquids in analytical chemistry. Anal. Chim. Acta, 2010, 661, 1-16. Bubalo, M.C.; Vidovic, S.; Redovnikovi, I.R.; Joki, S. Green solvents for green technologies. J. Chem. Technol. Biotechnol., 2015, 90, 1631-1639. Tu, G.R.; Liu, X.F.; Du, G.X.; Zhou, X.H.; Dang, H.J. Synthesis and characteristic determination of nanosized Fe3O4 powder. Fine Chem., 2004, 21 (9), 641-644.

Filik and Avan [36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

Yuanbi, Z.; Zumin, Q.; Jiaying, H. Preparation and analysis of Fe3O4 magnetic nanoparticles used as targeted-drug carriers. Chinese J. Chem. Engin., 2008, 16(3), 451-455. Srivastava, S.; Awasthi, R.; Gajbhiye, N.S.; Agarwal, V.; Singh, A.; Yadav, A.; Gupta, R.K. Innovative synthesis of citrate-coated superparamagnetic Fe3O4 nanoparticles and its preliminary applications. J. Colloid Interf. Sci., 2011, 359(1), 104-111. Nigam, S.; Barick, K.C.; Bahadur, D.; Development of citratestabilized Fe3O4 nanoparticles: Conjugation and release of doxorubicin for therapeutic applications. J. Magn. Magn. Mater., 2011, 323(2), 237-243. Vitouchová, M.; Janá, L.; Sommer, L. Interaction of iron(II) and the simultaneous spectrophotometric determination of Fe, Cu, Zn, Co and Ni with 2- (5-bromo-2-pyridylazo) -5-diethylaminophenol. Fresen. J. Anal. Chem., 1992, 343(3), 274-279. Ferreira, S.L.C.; de Jesus, D.S. The spectrophotometric and derivative spectrophotometricdetermination of cobalt in the presence of large amounts of nickel using Br-PADAP. J. Braz. Chem. Soc., 1996, 7(2), 109-114. Jiang, Z. -T.; Li, R.; Yu, J.C. Determination of cobalt in foods by -cyclodextrin polymer phase spectrophotometry using 2- (5bromo-2-pyridylazo) -5-diethylaminophenol. Anal. Lett., 2002, 35(5), 825-835. Ruiz, E.; Ferro, V.R.; Palomar, J.; Ortega, J.; Rodriguez, J. J. Interactions of ionic liquids and acetone: Thermodynamic properties, quantum-chemical calculations, and NMR analysis. J. Phys. Chem. B., 2013, 117(24), 73887398. Li, C.-H.; Gao, K.-X.; Meng, Y.-N.; Wu, X.-K.; Zhang, F.; Wang, Z.-X. Solution Thermodynamics of imidazolium-Based ionic liquids and volatile organic compounds: benzene and acetone. J. Chem. Eng. Data, 2015, 60(6), 1600-1607. Tao, Y.; Sun, D.-W. Enhancement of food processes by ultrasound. Crit. Rev. Food Sci. Nutr. 2015, 55(4), 570-594.