Extraction-Chromogenic System for Nickel(II) Based on 5-methyl-4-(2-thiazolylazo)resorcinol and Aliquat 336

A water−isobutanol extraction-chromogenic system for NiII, based on the azo dye 5-methyl-4-(2-thiazolylazo)resorcinol (MTAR; H2L) and the ionic liquid Aliquat 336 (A336), was studied. Under the optimal conditions (cMTAR = 2.0 × 10–4 mol dm–3, cA336 = 5.6 × 10–3 mol dm–3, pH 8.5 and extraction time t = 1 min), NiII is extracted as a ternary complex which can be represented by the formula (A336)2[Ni(L)2]. In the absence of A336, or in a slightly acidic medium, a binary complex, [Ni(HL)2], with an absorption maximum at λ = 548 nm and a shoulder at 590 nm is formed. The following extraction-spectrophotometric characteristics were determined at the above-mentioned optimal conditions: λmax (545 nm), molar absorptivity (5.0 × 104 dm3 mol–1 cm–1), Sandell’s sensitivity (1.2 × 10–3 μg cm–2), Beer’s law limits (0.05–3.1 μg cm–3), constant of extraction (Log K = 6.1) and fraction extracted (99.2%). The effect of foreign ions was studied; the most serious interferences were caused by CoII, CuII and CrIII.


Introduction
Nickel is a first-row transition metal with many industrial applications 1 and important functions in biology of microorganisms and plants. 2 It is involved in a variety of products, such as stainless steels, alloys, rechargeable batteries, guitar strings, green colored glasses, permanent magnets and catalysts for large-scale industrial processes. Most of the nickel produced is used in stainless steels, ferrous and non-ferrous alloys and nickel plating. 3 The chief criteria of use are its resistance to air oxygen, alkalis and acids, lustrous shine, ductility, malleability, and excellent ability to alloy with both ferrous and nonferrous metals. Over 3000 alloys containing nickel are known. They have more than 250 000 applications. 4 In the earth's crust, nickel occurs most often in combination with sulfur, iron and arsenic. It is found in all types of soils, in fossil fuels, marine sediments, volcanic emissions and iron meteorites. Anthropogenic activities that contribute to nickel loadings in the environment include burning of fuel and residual oil, mining, smelting, refining, alloy processing, scrap metal reprocessing, waste incineration and disposal of sewage sludge or application of sludge as a fertilizer. [4][5][6] It is known that nickel may have carcinogenic, mutagenic and allergic properties. 7 Negative health effects have been documented for respiratory, cardiovascular, gastrointestinal, hematological, musculoskeletal, hepatic, renal, dermal, ocular, immunological, developmental, neurological and reproductive systems. 6 An allergic skin reaction is a common problem for some people; it is often associated with the presence of nickel in daily-used objects, such as jewelry, keys, jean buttons, children's clothing and toys. 8,9 Many analytical techniques, such as atomic absorption spectrophotometry, optical emission spectrometry, X-ray fluorescence analysis, neutron activation analysis, voltammetry and spectrophotometry, have been used for Toncheva et al.: Extraction-Chromogenic System for Nickel(II) ... nickel determination. [10][11][12][13][14] The methods involving spectrophotometry are simple and low-cost; they can be easily combined with procedures for preliminary separation and concentration, such as cloud point extraction, 13 solid phase extraction 14 and liquid-liquid extraction (LLE). [15][16][17][18][19][20] 5-methyl-4-(2-thiazolylazo)resorcinol (MTAR) is a well-known analytical reagent. [21][22][23][24][25][26][27][28] It has recently been used in our laboratory for LLE of V IV,V . [29][30][31][32] Here, we describe investigations on a LLE-chromogenic system for Ni II containing MTAR and the ionic liquid Aliquat 336 (A336). The selected organic solvent was isobutanol (2-methylpropan-1-ol). It is characterized by a low toxicity, volatility and corrosivity, 33,34 can be produced from renewable resources, 35 fulfills the so-called "CHON principle" 36 and is readily biodegradable and non-bioaccumulative. 33,37

1. Reagents and Apparatus
Stock solution of Ni II (1.7 × 10 -2 mol dm -3 ; pH 2) was prepared by dissolving NiSO 4 7H 2 O (Sigma-Aldrich, 99.999% trace metals basis) in distilled water containing H 2 SO 4 . Working solutions (2 × 10 -4 mol dm -3 ) were prepared by appropriate dilution of the stock solution. MTAR (95%, Sigma-Aldrich Chemie GmbH) was dissolved in the presence of KOH; 31 the obtained slightly alkaline aqueous solution (pH 8-9) was at concentration of 2 × 10 -3 mol dm -3 . Aliquat 336 was purchased from Sigma-Aldrich Chemie and dissolved in isobitanol (p. a., Merck). Solutions with concentrations of 1.4 × 10 -2 and 5.6 × 10 -3 mol dm -3 were used; the calculations were based on the average molar mass of A336 (432 g mol −1 ). 38 The acidity of the aqueous medium was set by the addition of buffer solution, prepared by 2 mol dm -3 solutions of CH 3 COOH and ammonia. The pH of the buffers was measured by a WTW InoLab 7110 (Germany) pH meter with an accuracy of ±0.001 pH units. Absorbance measurements were made with a Camspec M508 and a Ultro-spec3300 pro UV-Vis spectrophotometers (UK), equipped with 1 cm path-length glass cells. Distilled water was used throughout the work.

Determination of the Optimum Conditions
Solutions of Ni II , MTAR and buffer were placed into a separatory funnel. The aqueous phase volume was adjusted to 5 cm 3 or 10 cm 3 with water. A portion of the A336 solution was added and the organic phase was adjusted to 5 cm 3 with isobutanol. Then the funnel was shaken for a fixed time interval (between 5 and 300 seconds). After separation of the phases, the aqueous layer was discarded, and the organic layer was transferred into a beaker. A pinch of anhydrous Na 2 SO 4 was added to remove any water and the colored solution was poured into the spectrophotometer cell. The absorbance was measured against isobutanol or simultaneously prepared blank sample.

3. Determination of the Distribution Ratio
The distribution ratio D was calculated by the formu- where A 1 is the absorbance obtained after a single extraction under the optimal pH and reagents concentrations (Table 1) and A 3 is the absorbance obtained after a triple extraction. 30,31 The single extraction and the first stage of the triple extraction were performed with equal volumes of both phases (5 cm 3 ). The organic layers were transferred into two 25-cm 3 calibrated flasks and the flask for the single extraction was brought to volume with the isobutanol solution of A336 (5.6 × 10 -3 mol dm -3 ). The second stage of the triple extraction was performed by adding a 5-cm 3 portion of the A336 solution to the aqueous phase which remained after the first stage. The third stage was performed in the same manner. The two successive organic layers were transferred to the flask with the isobutanol extract obtained after the first stage. The volume was brought to the mark with the A336 solution. Absorbances A 1 and A 3 were measured against corresponding blanks.

1. Absorption Spectra of the Isobutanol-Extracted Complexes
Spectra of isobutanol-extracted Ni−MTAR complexes in the presence (1, 3) or absence (2, 4) of A336 are shown in Fig. 1. The spectra 1 and 2 are alike (1: λ max = 545 nm; 2: λ max = 548 nm), however in absence of A336 (2) the absorbance is lower, and a shoulder at about 590 nm is observed. Such a shoulder is slightly visible in spectrum 1. There is a tendency for it to become more and more imperceptible with increasing pH and decreasing the MTAR concentration.
At lower pH values (4.7; spectra 3 and 4), nickel is extracted to a negligible extent regardless of the presence or absence of A336. The absorbance of the blanks in the absence of A336 (2' and 4') is lower than that obtained in the presence of A336 (1' and 3' , respectively). This can be explained by the formation of an extractable ion pair between the anion of the azo dye (HL − ) 31,39,40 and the cation of the ionic liquid (A336 + ).
nia-acetate buffer solutions, which have the best buffering action at pH 4.75 (the pK a of acetic acid) and 9.25 (the pK a of NH 4 + ) 41 were used in all experiments. In the presence of A336 (series 1), maximal absorbance was obtained in slightly alkaline medium. The sharp decrease at pH < 6 can be attributed to incomplete complex formation due to the predominance of doubly protonated MTAR species. 42 In the absence of A336 (series 2), maximal absorbance was recorded in the slightly acidic region (pH ca. 6.5 -7.0). It makes sense that the course of series 1 and series 2 is identical in the acidic region (up to pH ca. 6); probably the absorbance obtained in the presence of A336 is due to the simultaneous extraction of two complexes, Ni-MTAR-A336 and Ni-MTAR. However, with the increase in pH, the contribution of the ternary complex to the total absorbance becomes greater.
The decrease in absorbance at pH > ca. 9 (series 1) can be attributed to two factors: (i) presence of hydrolyzed Ni(OH) + species; 43 and (ii) a competitive extraction of the (A336 + )(HL − ) ion pair which reduces the effective concentrations of the reagents to a larger extent at higher pH values (see spectra 1' and 3' in Fig. 1).
One can conclude from Fig. 2, and from the above mentioned pK a value of NH 4 + (9.25), that the optimal pH range for the extraction of Ni II is 8.3−9.0. We performed our further investigations at pH 8.5.

2. Effect of pH
The effect of pH on the absorbance obtained in the presence or absence of A336 is shown in Fig. 2. Ammo-

Effect of the Reagents Concentrations, Volume of the Aqueous Phase and Shaking Time
The effect of the MTAR concentration on the absorbance is shown in Fig. 3 for two different sets of conditions. There is a maximum in both series; at high concentrations of MTAR, the absorption decreases. With a 2:1 volume ratio (series 2; 10 cm 3 aqueous phase / 5 cm 3 isobutanol phase), the apparent molar absorptivity is high- er. The optimal reagent concentration for this volume ratio appears to be c MTAR = 2.0 × 10 −4 mol dm −3 . It is considerably lower (by almost an order of magnitude) than the MTAR concentration required for the extraction of V V . 31 The effect of the A336 concentration on the absorbance is shown in Fig. 4. The optimal concentration is c А336 = 5.6 × 10 −3 mol dm −3 . It is also lower than that required for the extraction of V V ; 31 this indicates that the Ni II -MTAR-A336 complex is more stable than the corresponding V V complex.
The effect of shaking time was studied under the optimal pH and reagents concentrations (Table 1). It was found that 50 seconds are enough for a quantitative extraction of Ni II . To avoid accidental errors, induced by the combination of short shaking times and different shaking rates, we shook for 60 seconds in our further experiments.
(PAR = H 2 L = 4-(2-pyridylazo)resorcinol) exist, [Ni(HL) L]and [NiL 2 ] 2-. 49 In the presence of a quaternary ammonium salt (tetradecyldimethylbenzylammonium chloride; Q + Cl -), 50 two kinds of complexes can be extracted in chloroform, [Ni(HL) 2 ] . n(Q + Cl -) and (Q + ) 2 [NiL 2 ]; the second one has a higher molar absorptivity. Karipcin et al. isolated in solid state water-insoluble octahedral 1:2 binary complexes with PAR 47 or 4-(2-thiazolylazo)resorcinol (TAR); 48 they can be represented by the general formula [Ni(HL) 2 ]. The TAR complex is partially soluble in chloroform and readily soluble in alcohols; in ethanol, it shows an absorption maximum at 543 nm and a shoulder at 586 nm. 48 It is clear from our investigations (see Fig. 1, Fig 2 and  Fig. 4) that at least two different Ni II -MTAR isobutanol-extractable complexes exist: (i) a binary Ni-MTAR complex; it is extracted in the absence of A336 (Fig. 1, spectrum 1); and (ii) a ternary Ni-MTAR-A336 complex (Fig. 1, spectrum 2). The MTAR:Ni II and A336:Ni II molar ratios in the ternary complex were determined by the straight-line method of

4. Molar Ratios, Structure and Equations of Complex Formation and Extraction
It is known that thiazolylazo-and pyridylazo dyes form 1:1 44 or 1:2 45-48 complexes with Ni II . In aqueous medium, at least two different 1:2 Ni-to-PAR complex species Asmus 51 (Fig. 5) and the mobile equilibrium method 52 (Fig.  6). Its formation and extraction can be represented by Eq. 1.
As mentioned above, a complex with the same formula (which is insoluble in water and soluble in alcohols) has been prepared with TAR. 48 The MTAR complex obtained by us has similar spectral characteristics (an absorption maximum of 548 nm and a shoulder at about 590 nm), indicating that it probably has a similar structure. Most likely, both MTAR complexes, [Ni(HL) 2 ] and (A336 + ) 2 [Ni(L) 2 ], have an octahedral geometry which is common for Ni II centers. 47,48,53,54 Donor atoms of each of the tridentate MTAR units should be the o-hydroxyl oxygen (relative to the azo-group) and the two nitrogens (from the thiazole ring and azo-group). 47,48,55 As shown in Eqs. 1 and 2, MTAR at the operating conditions exists mainly in its HL − form (with a proton in the OH-group in ortho-position relative to the azogroup). 42,56 It is known that this form dominates (in aqueous medium) at the pH range of 5.7−11.8. 42

5. Beer's Law and Analytical Characteristics
The dependence between the concentration of Ni II in the aqueous phase and the absorbance of the extract was studied under the optimal conditions (Table 1). A good linearity was obtained in the range of 0.05 -3.1 µg cm -3 (R 2 = 0.9993, N = 9). The linear regression equation was A = 0.845γ -0.006, where A is the absorbance and γ is the concentration (μg cm -3 ) of Ni II . The standard deviations of the slope and intercept were 0.008 and 0.013, respectively. The limits of detection (LOD) and quantitation (LOQ), calculated as 3-and 10-times standard deviation of the intercept divided by the slope, were LOD = 46 ng cm -3 and LOQ = 153 ng cm -3 . The molar absorptivity (ε) and Sandell's sensitivity (S) at λ max = 545 nm were ε = 5.0 × 10 4 dm 3 mol -1 cm -1 and S = 1.2 × 10 -3 μg cm -2 , respectively.  The distribution ratio D was calculated from the absorbances obtained after single and triple extractions as described above: D = 121 ± 3 (six replicate measurements). The fraction extracted, E = 99.2%, was found by the formula E% = 100 × D/(D+1). The conditional equilibrium constant characterizing Eq. 1 was calculated by the mobile equilibrium method 52 (Fig. 5, straight line 2; abscissa intercept). The obtained value was Log K = 6.1 ± 0.2.

6. Effect of Foreign Ions
Various salts were used to test the selectivity of the developed procedure. Their effect is summarized in Table  2. The most significant interferences were those of Co II , Cu II and Cr III which form intensively colored complexes under the specified conditions. Cr III causes a negative error, while Co II and Cu II cause positive errors. Zn II also causes positive errors, however its interfering effect is smaller. A 250-fold excess of I − ; 100-fold excess of Mo VI ; 8.5-fold excess of Al III ; and 5-fold excess of V V , Fe III , Cd II and Mn II are tolerable. The determination of Ni II is not af-

Conclusions
The described investigations shed light on the complex formation between Ni II and MTAR in the presence or absence of A336. The conditions of existence of the two main extracted complexes, (A336 + ) 2 [Ni(L) 2 ] and [Ni(HL) 2 ], have been outlined and the differences in their spectra have been highlighted. In a slightly alkaline medium, Ni II predominantly forms a ternary complex. It is intensively colored and readily extractable into isobutanol. This solvent is less toxic than the diluents commonly applied to improve the physico-chemical characteristics of the ionic liquid А336 (e.g., benzene, 57 carbon tetrachloride, 58 chloroform, 58 toluene, 59 xylene 60 and kerosene 61 ) and the organic solvents typically used for LLE-spectrophotometric determination of Ni II (e.g., chloroform). [17][18][19][20]