2020 Volume 60 Issue 4 Pages 812-814
This paper suggests an improved method for the sample preparation to quantify vanadium in high-speed steel by using flame atomic absorption spectrometry. The suggested method prepared a solidified glass body fused with lithium tetraborate after the sample was decomposed with an acid mixture of hydrofluoric and nitric acids. The obtained borate glass was easily dissolved with nitric acid without any residues. Another novelty of this paper was a simultaneous measurement of a vanadium absorption line and a nickel absorption line in the close wavelength, when a multi-wavelength high-resolution spectrometer system was employed. The nickel line worked as an internal standard and could contribute to more precise quantification of vanadium in high-speed steel samples.
Vanadium is one of the alloyed elements in steel materials to improve the toughness and abrasion resistance. Therefore, the content should be determined with high precision for the quality control of manufactured materials as well as saving of the resource. The steel analysis for vanadium is now conducted with several analytical methods including flame atomic absorption spectrometry (AAS). However, the Japanese Industrial Standard (JIS) does not provide how vanadium should be determined in a steel sample containing tungsten,1) such as tool steel and high-speed steel, because it is co-precipitated with tungsten hydrolysis in a perchloric acid solution with the JIS recommended in the decomposition method. Further studies are thus needed to prepare the analyte solution for AAS in tungsten-containing steel materials.
Our previous paper has described an analysis of high-speed steel in which lithium borate fusion is carried out to prepare a nitric-acid solution in inductively coupled plasma-atomic emission spectrometry (ICP-AES).2) The prepared solution for a high-speed steel sample had an analytical stability without tungsten precipitation. This paper suggests a new method for the sample preparation, based on a solidified glass body fused with lithium tetraborate, in the quantification of vanadium by flame AAS. In addition, this paper reports on a simultaneous detection of the vanadium atomic line of 318.398 nm, which is the most sensitive line in AAS, and the nickel atomic line of 318.437 nm, when a multi-wavelength spectrometer equipped with a continuum light source was employed. The nickel line was measured as an internal standard,3,4) contributing to more reliable determination of vanadium in several high-speed steel samples.
The present quantification of vanadium was conducted using flame AAS. An atomic absorption spectrometer (contrAA 700; Analytik Jena AG, Jena, Germany) was used. This spectrometer comprises a xenon lamp as a continuum light source and an echelle grating system, which has an ability to measure a spectrum in a wavelength range of 0.2–0.3 nm. An optimal condition for the acetylene – nitrous oxide flame was predetermined as follows: the flow rate was 230 dm3 h−1, the observation height was 8 mm above the burner head, and the integration time for absorption signal was 3 s.
2.2. Reagents and SamplesAcid reagents were 70 mass% of nitric acid and 50 mass% of hydrofluoric acid. Borate fusion was conducted using lithium tetraborate as an alkaline flux. Tartaric acid was used as a complexing agent for tungsten in the nitric-acid specimen solution. A standard solution of vanadium was prepared for the calibration by digesting pure-vanadium metal with nitric acid. A standard solution of iron was prepared for the matrix matching by digesting pure-iron metal with nitric acid. An aluminum chloride solution containing 20 mg cm−3 of aluminum was prepared to add as a matrix modifier of vanadium atomization. A nickel sulfate (hexahydrate) solution containing 50 mg cm−3 of nickel was prepared to add an internal standard for the vanadium measurement.
Samples of high-speed steel were the following certified reference materials (CRMs) of the Japanese Iron and Steel Federation (Tokyo, Japan): JSS 607-9, JSS 609-11, and JSS 610-10.
2.3. Optimization of Co-existing AgentsTesting solutions containing several concentrations of constituent elements were prepared to optimize the AAS analysis. These solutions commonly contained 5.2 μg cm−3 of vanadium and 150 μg cm−3 of iron as the major component in a steel sample. Three sets of testing solutions were prepared as the following: (a) 0, 2.5, 5.0, 7.5, and 10 mg cm−3 of lithium tetraborate as a flux agent together with 2.0 mg cm−3 of aluminum; (b) 0, 1.0, 2.0, 3.0, and 4.0 mg cm−3 of aluminum (added by aluminum chloride solution) as a matrix modifier together with 5.0 mg cm−3 of lithium tetraborate; and (c) 0, 2.5, 5.0, 10.0, and 15.0 mg cm−3 of nickel (added by nickel sulfate solution) as an internal standard together with 5.0 g cm−3 of lithium tetraborate and 2.0 mg cm−3 of aluminum.
2.4. Solution PreparationA high-speed steel sample (0.050 g) was placed in a platinum-5 mass% gold-alloyed vessel, which was used to prepare glass bead specimen on X-ray fluorescence analysis.5,6) An acid mixture (about 0.6 cm3) of hydrofluoric and nitric acids was added to the steel sample and heated on a hot-plate (about 450 K). The black-colored residue left after the acid digestion was melted with 1.0 g of lithium tetraborate using a Meker burner. Borate glass, which was obtained by vitrifying the melted mixture, was dissolved with about 40 cm3 of an acid mixture containing 1.0 g of tartaric acid, 2 cm3 of nitric acid, and distilled water. The solution was placed in a 100-cm3 of volumetric flask and diluted with 8 cm3 of nitric acid and distilled water. After then, 25 cm3 of the prepared solution was divided and diluted to 50.0 cm3 with an addition of 0.25 g of tartaric acid, 5.0 cm3 of aluminum chloride solution, and 5.0 cm3 of nickel sulfate solution.
Seven calibrating standard solutions were prepared. These solutions contained vanadium of 0, 1.37, 2.75, 5.50, 8.24, 11.0, or 13.7 μg cm−3. They also contained the same amounts of lithium tetraborate, tartaric acid, aluminum, and nickel as those in the sample solution.
The condition for the acetylene – nitrous oxide flame was tested as shown in Fig. 1. Although the higher absorbance of vanadium was obtained by using more than 240 dm3 h−1 of the flow rate, the formation of soot during the AAS measurement might prevent the precise and accurate quantification. Therefore, 230 dm3 h−1 of the flow rate of acetylene was employed.
Variations in the absorbance of vanadium and nickel absorption lines with the flow rate of acetylene gas. ●, V 318.398 nm for the analytical line; and ○, Ni 318.437 nm for an internal standard.
Solution preparation with borate fusion for ICP-AES2) was improved for the present AAS of vanadium quantification. Lithium tetraborate was used as an alkaline flux. This reagent might influence on a spectrochemical analysis. For example, it was reported that it largely decrease the emission intensity of Mg, Al, and Ca on ICP-AES measurement.7) Although this reagent had little effect on the vanadium atomization as shown in Fig. 2(a), the burner slit was blocked by precipitates of lithium tetraborate at the higher concentration during the repeated measurements. Thus, a specimen solution containing 5.0 mg cm−3 of lithium tetraborate was used. When 50 mg of the high-speed steel sample was decomposed with an acid mixture of hydrofluoric and nitric acids, lithium tetoraborate was required at least 1.0 g. Thus, 100 cm3 of a sample solution containing 1.0 g of lithium tetraborate was prepared, and then, 25 cm3 of this solution was diluted to 50 cm3 to obtain a specimen solution having 5.0 mg cm−3 of lithium tetraborate.
Variations in the absorbance of vanadium (318.398 nm) absorption line with additional reagents of (a) lithium tetraborate as an alkaline flux, (b) aluminum as a matrix modifier, and (c) nickel as an internal standard.
The absorbance of vanadium could be enhanced by adding aluminum in the specimen solution. A suitable concentration of aluminum for the present analysis was examined as shown in Fig. 2(b). The solution containing 2.0 mg cm−3 of aluminum was preferred.
3.4. Internal Standard of NickelIn the atomic absorption spectrum, the peak of nickel absorption (318.437 nm) appeared just near that of vanadium (318.398 nm) as shown in Fig. 3. Variations of their absorbances were checked by calculating a normalized absorbance, which was obtained from {(each absorbance) – (mean value)}/(standard deviation), as shown in Fig. 4. The absorbance of nickel was similarly varied along with the absorbance of vanadium. Thus, nickel was suitable for an internal standard in the sample and standard solutions. A suitable concentration of nickel was tested as shown in Fig. 2(c). Nickel sulfate solution was added to a specimen solution, and then, higher concentrations of nickel weakened the absorbance of vanadium. This might be because physical interference was caused by adding a large volume of the sulfate solution. The solution containing 5.0 mg cm−3 of nickel was preferable.
Atomic absorption spectrum (the wavelength range from 318.222 nm to 318.572 nm) of a specimen solution containing high-speed steel sample of JSS 609-11 and nickel as an internal standard.
Variations in the normalized absorbance of vanadium and nickel absorption lines with the number of continuous measurement for about 30 s. ●, V 318.341 nm; ▲, V 318.398 nm for the analytical line; ■, V 318.540 nm; ○, Ni 318.437 nm for an internal standard.
A high-speed steel sample contains about sub-mass% of nickel as an alloyed element. When no nickel internal standard was added, nickel concentrations in the present specimen solution would be 0.15 μg cm−3 for JSS 607-9 (certified value of nickel is 0.0603 mass%), 0.31 μg cm−3 for JSS 609-11 (certified value of nickel is 0.1234 mass%), and 0.22 μg cm−3 for JSS 610-11 (certified value of nickel is 0.087 mass%). On the other hand, the prepared specimen solution contained nickel at least 15000-times as much as the concentration in the CRMs. Therefore, the alloyed nickel in the high-speed steel could be ignored in this internal standard method.
3.5. Validation AnalysisCertified reference materials of high-speed steel were analyzed, and the obtained results were well agreed with the certified values (Table 1). Furthermore, the addition of nickel as the internal standard was very useful to obtain accurate results. A temporal variation of the flame condition (day of B as shown in Fig. 5) could be successfully corrected with the absorbance of nickel as the internal standard.
| CRMs | Without internal standard method | With internal standard method | Certified value | ||
|---|---|---|---|---|---|
| Day of A | Day of B | Day of A | Day of B | ||
| JSS 607-9 | 0.82 ± 0.02 | 1.13 ± 0.17 | 0.83 ± 0.02 | 0.85 ± 0.02 | 0.90 |
| JSS 609-11 | 1.71 ± 0.07 | 2.51 ± 0.20 | 1.78 ± 0.03 | 1.85 ± 0.03 | 1.87 |
| JSS 610-10 | 3.00 ± 0.26 | 4.56 ± 0.04 | 3.15 ± 0.08 | 3.25 ± 0.06 | 3.25 |
Variations in the absorbance of nickel absorption line (318.437 nm) with the number of continuous measurement for about 20 min on the quantification of vanadium in high-speed steel. Number of measurement: 1 to 7, calibrating standard solutions; and 8 to 16, sample solutions.
A sample solution of high-speed steel without tungsten precipitation could be prepared with an alkaline fusion of lithium tetraborate when it was analyzed using flame AAS. By adding nickel as an internal standard to the sample solution, the vanadium quantification could be conducted with higher precision by simultaneously measuring the absorbances of vanadium and nickel with a multi-wavelength high-resolution spectrometer system. The internal standard method was successfully applied to determine the vanadium content in several CRMs of high-speed steel.
