2015 Volume 55 Issue 1 Pages 332-334
Determination of trace amounts (~0.05 μg) of bismuth in steel by inductively coupled plasma mass spectrometry (ICP-MS) has been carried out by a cascade-preconcentration and separation method that combines the one-drop solvent concentration method and solid-phase extraction method. The bismuth(III) ion from acid decomposition of the steel sample formed a complex with iodine, and this subsequently formed an ion pair with tetrabutylammonium (TBA). This ion pair was almost completely adsorbed by passage through a solid-phase packed column. The ion pair was eluted with ethyl acetate, and a small volume of dimethyl sulfoxide (DMSO) was added to the resulting solution; successful preconcentration of the iodobismuthate complex/TBA ion pair into the DMSO phase was accomplished by removing the ethyl acetate using a compact evaporator in a water bath (40°C). The sample was diluted with an internal standard (thallium) solution and an acidic solution and was then tested by ICP-MS. The calibration curve for the proposed method showed good linearity (r2 = 0.9997) in the 0.05–10 μg range. A recovery test was performed, in which 0.05 μg and 2.5 μg of bismuth was spiked into an acidic decomposition solution of pure iron (JSS 003-6), and the recovery was 98% for both solutions, and the relative standard deviations (n = 5) were 8.8% and 1.2%, respectively. Good results were obtained applying the method to analysis of CRMs.
In steel, trace metals with low melting points, such as Bi, Sb, Pb, Sn, and As, can lead to deterioration in hot workability, strength, stability, and corrosion resistance. Quantification of these metals is carried out by methods including glow discharge mass spectrometry,1) anodic stripping voltammetry,2) graphite furnace atomic absorption,3) inductively coupled plasma optical emission spectrometry (ICP-OES),4) ICP mass spectrometry (ICP-MS),5) and flame atomic absorption spectrometry.6) Bismuth, in particular, has been widely studied. Oguma et al.7) developed a method for the determination of bismuth in steel by ICP-OES. This method involves the formation of a stable complex anion of ionic bismuth and iodide which is subsequently separated from the matrix using a strongly basic anion exchange resin. This is followed by elution with nitric acid and ICP-OES. Uehara et al.8) developed a method for the determination of bismuth complexes with ethylene diamine tetraacetic acid by high performance liquid chromatography.
Recently, the API standard (API6A718) was set at less than 0.5 ppm bismuth in the material standard of Alloy718 (nickel alloy); therefore, to determine levels with high sensitivity, it is necessary to separate trace levels of bismuth from matrix components and concentrate them.
In this study, we combined two different concentration and separation methods. Specifically, an iodobismuthate complex was formed between the bismuth ion and iodide. The bismuth complex was separated from the matrix component by ion-pair extraction using tetrabutylammonium bromide (TBA·Br), and the complex was adsorbed on a reverse-phase column. Heterogeneous preconcentration and a separation system9) (cascade-type10)) was used to recover the ion pair using ethyl acetate (b.p. 77°C); the sample was concentrated with a micro volume of DMSO (b.p. 189°C). Ethyl acetate plays an important role linking the two separation methods, as it was used as the eluent for the ion pairs in the solid-phase extraction method and was evaporated in the one-drop solvent concentration method. Thus, separation from the matrix with high concentration was achieved, and the sample was analyzed by ICP-MS to determine trace levels of bismuth.
Potassium iodide, TBA·Br, L(+)-ascorbic acid, and ethyl acetate (all special grade) were obtained from Wako Pure Chemical Industries, Ltd. DMSO (special grade) was obtained from Kanto Kagaku. Bismuth standard solution (Japan Calibration Service System (JCSS) standard solutions, 1000 mg/L) was obtained from Wako Pure Chemical Industries, Ltd. JSS 003-6 was used as a pure iron standard; NCS HC 11527, BCS 346, and NIST SRM 363 were used as certified reference materials. Water was purified with a Milli-RX12α ultrapure water system (Millipore Corp.). The solid-phase column (NOBIAS RP-OD1L) packed with silica modified with the octadecylsilyl group was obtained from Hitachi High Technologies. Filter paper (grade 5C, Toyo Roshi Kaisha, Ltd.) was used for filtrations. All other reagents were special grade and were commercially available, unless stated otherwise.
A BM-41 water bath (Yamato Scientific Co., Ltd.) and a RAPID EXTEST Single Flex G2 compact evaporator (Bio Chromato Inc.) equipped with an MDA-015 vacuum pump (ULVAC KIKO Inc.) were used for the concentration. The evaporated ethyl acetate was recovered using an aggregation cooling trap. Solid-phase extraction was automatically performed using an SPE-100 solid-phase extraction device (Hiranuma Sangyo Co. Ltd.). ICP-MS was performed using an ELAN DRC II system (Perkin-Elmer Inc.). Absorbance was measured using a Hitachi U-3310 UV–vis spectrophotometer. pH was measured using an F-51 pH meter (HORIBA Ltd.) equipped with a 9611-10D glass electrode (HORIBA Ltd.). The addition of the reagent solution was performed using a microsyringe (Hamilton Corp., 100 μL).
The sample (0.5 g) was placed in a 200 mL beaker and was covered with a watch glass. Mixed acid (nitric acid and hydrochloric acid, 50 mL) was added, and the mixture was gently heated to dissolve the sample. The solution was cooled, the watch glass was removed, and the bottom of the watch glass was washed with water. In cases where residue was present, the residue was filtered off (5C filter paper). L(+)-ascorbic acid solution (200 g/L, 5 mL) was added to the solution. The mixture was stirred for about 5 min. Potassium iodide solution (664 g/L, 5 mL: Molar ratio against bismuth is around 400000.) was added with gentle stirring, and the iodobismuthate complex was formed. Subsequently, TBA·Br solution (15.42 g/L, 1 mL: Molar ratio against bismuth is 1000.) was added to the gently stirred solution, and an ion pair of the iodobismuthate complex and TBA was formed. The solution containing the ion pair was passed through the solid-phase packed column to separate the ion pair from the matrix components (such as iron, chromium, and nickel) by adsorption on the column. Five percent nitric acid solution (5 mL) was passed through the column for washing, and air was passed through to remove moisture. Ethyl acetate (5 mL) was passed through the column to elute the ion pair, and the eluent containing the ion pair was placed in a centrifuge tube. A small volume of DMSO (50 μL) was added, and a compact evaporator was attached. The ethyl acetate was removed by evaporation for 15 min in a water bath at 40°C. The tube was evacuated at a pumping speed of 12 L/min using a vacuum pump. Thus, the ion pair was concentrated in the DMSO phase. The sample was diluted by addition of internal standard (thallium) solution and 5% nitric acid solution to the gently stirred concentrated DMSO phase. The sample was allowed to stand and cool to room temperature. The sample was diluted to 250 μL with 5% nitric acid solution. An aliquot of the solution was determined by ICP-MS using thallium as an internal standard element.
The effect of pH on the iodobismuthate complexation in the pH range 1.0–10.0 is shown in Fig. 1. A bismuth solution was placed in a 20 mL glass vial, and buffer solutions (pH 1.0, 2.0: HCl/KCl buffer; pH 3.0–6.0: acetic acid/sodium acetate buffer; pH 7.0: NH3/NH4Cl buffer; pH 8.0–10.0: borax buffer) were then added to control the pH. Potassium iodide solution (664 g/L, 5 mL) was added, and the sample was diluted to 20 mL with water. An iodobismuthate complex was formed by gently stirring the mixture. An aliquot of the solution was placed in a quartz cell, and the absorbance at 465 nm was measured using a UV–vis spectrophotometer. The iodobismuthate complex was stable when the pH was below 5.0. The acidity for decomposition of steel is usually around pH 1.0. This pH was used for further experiments without adjustment.
Effect of pH on iodobismuthate complexation as measured by UV-vis spectrometry. [Bi3+] = 0.5 μg mL−1 and [KI] = 166 mg mL−1.
The effect of the molar amount of iodine on the iodobismuthate complexation process was examined. Sufficient nitric acid was added to the sample to give a 5% nitric acid in bismuth (10 μg: 0.0479 μmol) solution; L(+)-ascorbic acid solution (200 g/L, 5 mL) was added. The appropriate amount of potassium iodide solution was added in the range of 0.1–40000 μmol. The iodobismuthate complex was formed and diluted to a constant volume with water. The absorbance at 465 nm was measured by UV–vis spectroscopy. The absorbance increased with the amount of potassium iodide added and became practically constant at 10000 μmol and above. From this result, the optimum concentration of potassium iodide was 20000 μmol. In this added amount, the existence percentage of iodobismuthate (BiI63–) was estimated at 99.0% by a calculation with the stepwise stability constant.11)
The effect of the molar ratio of TBA/Bi on the ion pair formation was examined. The iodobismuthate complex was formed by the procedure described in the previous section. The amount of TBA added was in the range 50–2000 times (molar ratio) that of bismuth to form the ion pair. The solution containing the ion pair was passed through the solid-phase packed column, and the ion pair was adsorbed on the stationary phase. Five percent nitric acid solution (5 mL) was passed through the column for washing, and air was passed through to remove moisture. Ethyl acetate (5 mL) was passed through the column to elute the ion pair, and the eluent was placed in a 200 mL beaker. The sample was digested with a mixture of sulfuric acid and nitric acid and diluted with water to a constant volume of 100 mL. The diluted solution was measured by ICP-MS. Almost 100% of the bismuth was recovered when a molar ratio of 500 or more was employed. As a result, a molar ratio of 1000 times, i.e., 15.42 mg of TBA·Br, was selected as the optimum condition.
The adsorption-elution results, expressed as percentages, for the ion pair in the solid phase extraction was examined. The ion pair of the iodobismuthate complex and TBA was formed under the optimum conditions. The solution containing the ion pair was passed through the solid-phase column, and the residual solution containing non-adsorbed complex was decomposed with nitric acid. The solution was diluted with water to a constant volume of 100 mL. The dilute solution was measured by ICP-MS, and the adsorption percentage was calculated from the results. Subsequently, 5% nitric acid solution (5 mL) was passed through the column for washing, and air was passed through to remove moisture. Ethyl acetate (5 mL) was passed through the column to elute the ion pair, and the eluent was placed in a 200 mL beaker. The sample was digested with a mixture of sulfuric and nitric acids and was diluted with water to a constant volume of 100 mL. The dilute solution was measured by ICP-MS, and the elution percentage was calculated from the results. The values obtained, i.e., 99.1% adsorption and 99.9% elution indicated that the ion pair between the iodobismuthate complex and TBA was almost completely adsorbed on the solid-phase packed column and was almost entirely recovered by elution with ethyl acetate (5 mL).
Previous studies12) for the removal of ethyl acetate and DMSO with a compact evaporator reported that almost 100% of the ethyl acetate can be removed by evaporation within 15 min, whereas DMSO barely evaporated at all under these conditions. When the optimum evaporation time of 15 min was applied in the process reported here, 99.6% of the ion pair had moved to the DMSO phase. Therefore, during the evaporative removal of ethyl acetate over 15 min, the ion pair had moved to the DMSO phase and was concentrated to a micro volume (50 μL).
Under the optimum conditions, a calibration curve for bismuth was created. The calibration curve showed good linearity (r2 = 0.9997) for bismuth in the 0.05–10 μg range. In addition, the lower limit of quantitation (10σ) was 0.01 μg/g (R.S.D. = ~13%) for a sample weight of 0.5 g.
Results of analysis of certified reference materials (nickel alloy and Cr–V steel) and a recovery test using pure iron with the proposed method are shown in Table 1. In the recovery test, 0.05 μg and 2.5 μg of bismuth was spiked into an acidic decomposition solution of pure iron (JSS 003-6), and the recoveries were 98% for both, and the relative standard deviations (n = 5) were 8.8% and 1.2%, respectively. Against certified values for BCS 346 (10.4 ± 0.4 μg/g) and NCS HC 11527 (1.2 ± 0.3 μg/g), values of 10.1 ± 0.2 μg/g and 1.1 ± 0.1 μg/g were obtained, respectively.
A 200-fold increase in concentration (i.e., 50 mL→250 μL) was achieved by a cascade preconcentration and separation approach, in which two different preconcentration methods were combined, namely a solid-phase extraction method (solid-phase column packed with silica modified with an octadecylsilyl group) and the one-drop solvent concentration method, using ethyl acetate and DMSO. The proposed concentration and separation system was effective for the determination of sub-ppm levels of bismuth in steel.