2024 Volume 64 Issue 10 Pages 1615-1617
This paper reports an improved method for sample preparation of a stainless steel sample for inductively coupled plasma mass spectrometry. Conventional digestion methods using a digestion agent containing hydrochloric acid affect chlorine spectral interference such as that by 35Cl16O+ or 40Ar35Cl+. Alternatively, a microwave digestion method using an acid mixture of nitric acid and hydrofluoric acid can fully eliminate these interferences. The suggested procedure can contribute to more reliable quantification of titanium, vanadium, arsenic, niobium, tin, and antimony in stainless steel samples using a low-resolution quadrupole mass spectrometer.
Trace elements in a steel product can strongly affect its physical and chemical properties including resistance to abrasion, impact, corrosion, heat, and weather. Therefore, accurate and reliable chemical analyses must be conducted for quality control. Inductively coupled plasma mass spectrometry (ICP-MS) can be used frequently for this purpose. Although pretreatments of matrix iron separation (e.g., chelate separation,1,2) polyurethane form purification,3) 4-methyle-2-pentanone extraction,4) ion exchange,5,6) cascade-preconcentration,7) and solid phase extraction resin8)) are preferred for ICP-MS analysis, direct measurement of a digested steel sample is also an important technique.9) Quadrupole mass spectrometers are generally used for ICP-MS analysis. Analysts have been confronted with the difficulty of spectral interference such as 32S16O+ for 48Ti+, 35Cl16O+ for 51V+, and 40Ar35Cl+ for 75As+. Interference of these types results from co-existing acid reagents of sulfuric acid (32S) and hydrochloric acid (35Cl) in a specimen solution. For example, the following collision and reaction methods4,8) have been used to overcome these interferences: (1) helium collision to decompose interfered polyatomic ions, (2) oxygen reaction to measure a generated polyatomic ion of an analyte (e.g., 48Ti16O+ for titanium quantifications, 51V16O+ for vanadium quantifications, and 75As16O+ for arsenic quantifications), and (3) hydrogen reaction to inhibit argon polymetric ions (i.e., 40Ar35Cl+). Such collision and reaction methods are not completely adequate that they are still uncertain.
Usually, stainless steel is digested using an acid mixture of hydrochloric and nitric acids10) or a sulfuric acid agent.11) For example, the author has encountered some acidic mixtures used for microwave digestion of various iron and steel samples: (i) hydrofluoric, nitric, and phosphoric acids;12,13,14,15) (ii) hydrochloric, hydrofluoric, nitric, and phosphoric acids;12) and (iii) hydrochloric, hydrofluoric, nitric, and sulfuric acids.16) However, stainless steel samples require hydrochloric acid.12,16) Then, ICP-MS quantification based on these digestion techniques must be conducted by considering the application of co-existing chlorine (35Cl and 37Cl), phosphorous (31P), and sulfur (32S, 33S, 34S, and 36S). In this work, microwave digestion using nitric-acid agents proposed for stainless steel samples. A microwave oven can enable nitric acid to digest stainless steel samples. The analytical procedure developed in this study leads to an interference free determination of several minor and trace elements (titanium, vanadium, arsenic, niobium, tin, and antimony) without collision/reaction cell technique.
A microwave oven (ETHOS UP; Milestone srl., Sorisole, Italy) was used to digest stainless steel samples. Also, a tandem quadrupole mass spectrometer of iCAP TQ (Thermo Fisher Scientific Inc., Waltham, MA, USA) was used for quantitative ICP-MS measurements. The instrument conditions are presented in Table 1. A water purifier of Milli-Q Direct 8 (Merck KGaA, Darmstadt, Germany) was used to prepare > 18 MΩ cm purified water.
| Radio frequency of generator | 27.12 MHz | |
| Power of generator | 1.55 kW | |
| Flow rate of argon gas | Coolant gas | 14 dm3 min−1 |
| Auxiliary gas | 0.8 dm3 min−1 | |
| Nebulizer gas | 0.877 dm3 min−1 | |
| Nebulizer | Perfluoroalkoxy alkane concentric type | |
| Spray chamber | Perfluoroalkoxy alkane cyclone type | |
| Chamber temperature | 275.85 K | |
| Injector material | Sapphire | |
| Injector size | 2.0 mm inner diameter | |
| Sampling and skimmer cone material | Platinum | |
| Peristaltic pump rate | 0.4 cm3 min−1 | |
| Accumulation time | 0.1 s | |
| Measured ion (internal standard) | Titanium | 48Ti+ (45Sc+) |
| Vanadium | 51V+ (89Y+) | |
| Arsenic | 75As+ (89Y+) | |
| Niobium | 93Nb+ (89Y+) | |
| Tin | 118Sn+ (89Y+) | |
| Antimony | 121Sb+ (89Y+) |
The following certified reference materials (CRMs) of stainless steel were used: JSS 651-16, JSS 652-15, JSS 653-16, and JSS 654-16 (The Japanese Iron and Steel Federation, Tokyo, Japan); and ECRM 231-2 (Bundesanstalt für Materialforschung und -prüfung, Berlin, Germany). The CRMs of JSS were used for digestion test using 7.8 mol dm−3 nitric acid. The three CRMs of JSS 651-16, JSS 654-16 and ECRM 231-2 were used for validation of this analytical work.
The following acids were used for steel digestion: acid reagents for the electronics industry (Fujifilm Wako Pure Chemical Corp., Osaka, Japan) of 35 mass fraction % hydrochloric acid and 70 mass fraction % nitric acid, and electronic grade reagent (Morita Chemical Industries Co., Ltd., Osaka, Japan) of 50 mass fraction % hydrofluoric acid.
After a stainless steel sample (0.10 g) was weighed using an electronic balance, it was placed in a 100-cm3 polytetrafluoroethylene digestion vessel. The following two digestion agents were prepared: (1) 7.8 mol dm−3 nitric acid; and (2) a mixture containing 7.8 mol dm−3 nitric acid and 0.54 mol dm−3 hydrofluoric acid. The prepared agent (5 cm3) was poured in the vessel, which was set on the microwave oven. The oven was operated using the following heating program: increasing the temperature from 298 K to 513 K in 15 min and maintaining this increased temperature for 120 min. After cooling the vessel to less than 313 K, the resulting solution was placed in a 25-cm3 polypropylene volumetric flask. By diluting this solution, specimen solutions for the ICP-MS measurement were prepared. Prepared specimen solutions consisted of 0.10, 0.08, or 0.06 mg cm−3 digested steel; about 0.3 mol dm−3 nitric acid; about 0.003 mol dm−3 hydrofluoric acid (for the use of agent (2)); and 4 ng cm−3 scandium and yttrium as internal standard elements. A fluoride compound of rare earth elements is readily precipitated in a hydrofluoric-acid solution because of its low solubility. However, 4 ng cm−3 scandium and yttrium were applicable to the present solution preparation, as presented in Fig. 1. When small amounts of these rare earth elements are added to a hydrofluoric-acid specimen solution, they would not precipitate for at least 7 days. Similarly, an yttrium internal standard of ICP-MS had been used for an International Organization for Standardization (ISO) method9) of hydrofluoric-acid use. These solutions containing varied amounts of steel were prepared to conform the robustness of the present calibration for overcoming elemental fractionation effects. If an unsuitable calibration is adopted, then accurate analytical results would not be obtained.
, 4 ng cm−3 yttrium; 
, 4 μg cm−3 yttrium; 
, 4 ng cm−3 scandium; and 
, 4 μg cm−3 scandium. (Online version in color.)Metal standard solutions except for arsenic were prepared by digesting each pure metal (at least 99.9 mass fraction % or purer). A 1.0 mg cm−3 non-certified standard solution of arsenic (Certipur 119773; Merck KGaA, Darmstadt, Germany) was purchased. Stock solutions of scandium and yttrium were prepared by digesting each oxide (at least 99.9 mass fraction % or purer). Two sets of calibrating standard solutions were prepared: (1) no matrix metals, and (2) 0.10 mg cm−3 of iron (99.99 mass fraction % pure iron) as a matrix constituent. When an internal standard method was applied successfully to obtain the almost identical slopes of calibration curve between (1) and (2), the calibration curve obtained using the former standards of (1) was used to estimate analytical results for the present ICP-MS measurements. This method obviated the confirmation of analyte impurities from the analytical procedure.
Stainless steel can be digested easily and rapidly using an acid mixture of hydrochloric and nitric acids. By contrast, nitric acid does not digest stainless steel using hot-plate heating because of the chromium passive film on the steel surface. However, microwave heating with 7.8 mol dm−3 nitric acid was applied successfully to digest a stainless steel sample, as shown in Table 2. Chromium-rich stainless steels (> 23 mass fraction %) required a higher temperature (513 K) than stainless steels containing lower contents of chromium (< 19 mass fraction %), which could be digested at 493 K. None of the tested stainless steels was digestible by the microwave heating at 473 K. The chromium passive film might not form at 513 K. Similarly pure chromium metal (99.99 mass fraction %; 0.10 g) was digestible using 7.8 mol dm−3 nitric acid (10 cm3) at 513 K for 600 min of cumulative total heating time.
| Sample name | Chromium content/mass fraction% | Maximum temperature on microwave heating | ||
|---|---|---|---|---|
| 473 K | 493 K | 513 K | ||
| JSS 651-16 | 18.12 | Undigested | Digested | Digested |
| JSS 652-15 | 16.86 | Undigested | Digested | Digested |
| JSS 653-16 | 23.18 | ― | Undigested | Digested |
| JSS 654-16 | 24.92 | ― | Undigested | Digested |
The analytical values for the three types of CRMs obtained by this method are shown in Table 3, along with the results and certified values obtained by the ISO method. They were generally close to the certified values. Although low values were given for some elements (Nb, Sn, and Sb) when dissolved with nitric acid alone, this was thought to be due to hydrolysis, suggesting the validity of adding hydrofluoric acid. Although specimen solutions of the ISO method9) using 3 cm3 hydrochloric acid, 1 cm3 nitric acid, and 0.5 cm3 hydrofluoric acid were also prepared, these solutions yielded erroneous results for vanadium and arsenic, as shown in Table 3. The two analytical values improved when the collision reaction method using oxygen gas in the ISO method, which revealed that the molecular ion interference was caused by chlorine. Lower limits of quantification related to 10 times the standard deviation of blank measurements were 0.2 mg kg−1 titanium, 0.02 mg kg−1 vanadium, 0.2 mg kg−1 arsenic, 0.03 mg kg−1 niobium, 0.02 mg kg−1 tin, and 0.05 mg kg−1 antimony.
| Sample name | Analyte | ISO digestion | Present digestion | Certified value (uncertainty) | |
|---|---|---|---|---|---|
| Using HCl, HNO3, and HF | Using HNO3 | Using HNO3 and HF | |||
| ECRM 231-2 | Titanium | 6.3 ± 0.2 | 6.3 ± 0.1 | 6.2 ± 0.1 | 7 (2) |
| Vanadium | 981 ± 66 | 699 ± 13 | 696 ± 3 | 708 (8) | |
| Arsenic | 157 ± 27 | 46.1 ± 0.7 | 46.1 ± 0.3 | 48 (3) | |
| Niobium | 23.5 ± 3.2 | 20.3 ± 0.3 | 20.3 ± 0.0 | ― | |
| Tin | 42.7 ± 0.2 | 40.5 ± 1.2 | 41.7 ± 0.3 | 43 (3) | |
| Antimony | 10.7 ± 0.0 | 8.4 ± 1.3 | 10.6 ± 0.1 | 11 (1) | |
| JSS 651-16 | Titanium | 16.2 ± 1.4 | 12.5 ± 0.2 | 14.1 ± 0.1 | ― |
| Vanadium | 1028 ± 59 | 751 ± 12 | 731 ± 7 | 746 (13) | |
| Arsenic | 151 ± 28 | 40.5 ± 0 | 40.1 ± 0.3 | ― | |
| Niobium | 166 ± 7 | 33.9 ± 2.2 | 155 ± 1 | ― | |
| Tin | 119 ± 1 | 17.5 ± 7.6 | 118 ± 1 | ― | |
| Antimony | 11.2 ± 0.1 | 0.2 ± 0.1 | 10.7 ± 0.1 | ― | |
| JSS 654-16 | Titanium | 16.5 ± 0.4 | 15.6 ± 0.3 | 15.8 ± 0.3 | ― |
| Vanadium | 814 ± 51 | 542 ± 13 | 533 ± 8 | 552 (12) | |
| Arsenic | 125 ± 26 | 24.6 ± 0.2 | 24.2 ± 0.5 | ― | |
| Niobium | 47.7 ± 1.3 | 33.5 ± 1 | 44.6 ± 0.6 | ― | |
| Tin | 23.5 ± 0.5 | 12.4 ± 2.9 | 22.4 ± 0.5 | ― | |
| Antimony | 5.1 ± 0.1 | 1.1 ± 0.9 | 4.8 ± 0.1 | ― | |
Determination of titanium, vanadium, arsenic, niobium, tin, and antimony in steel was successful using nitric acid and hydrofluoric acid as decomposing reagents. By performing pretreatment without hydrochloric acid, it was possible to remove molecular ion interference caused by chlorine, thus enabling to design a simple ICP-QMS method without collision reaction cell technique.
