The determination of trace elements in iron and steel requires separation procedures to obtain the accurate and precise analytical results. The present review covers recent advances in separation techniques, including sorption, liquid–liquid extraction, precipitation, electrolysis, and volatilization.
A novel flow injection (FI) system is presented for simple, rapid and sensitive determination of cadmium at μg/g to sub-μg/g levels in iron and steel. The anion-exchange column incorporated in-line in a FI system was utilized as an integrated media for both the separation/preconcentration of cadmium from a large excess of iron matrix and the detection reaction of adsorbed cadmium with TMPyP (5,10,15,20-Tetrakis(N-methylpyridinium-4-yl)-21H,23H-porphine,tetrakis(p-toluene-sulfonate)) which was directly introduced into the column by switching its flow. The spectrophotometric detection at 512 nm for Cd complex with TMPyP was directly coupled in-line with such separation/preconcentration and reaction in this FI system which consists of a multi-position valve and a syringe pump and is operated automatically under the control by a personal computer. The variables relating to such reactions and manifold were studied in detail to establish the optimum conditions and manifold configulations. A linear calibration using a 400 μL sample injection was obtained for cadmium in the range of 0–20 μg/L (in the presence of 1.0×104 μg/mL iron matrix). The coefficient of variation for 10 μg/L cadmium was 1.8% (n=3), and the estimated limit of determination was 1 μg/L corresponding to 0.05 μg/g (5×10−6 mass%) in iron and steel sample. Only 14 min was required for an analytical procedure. Satisfactory recoveries of spiked cadmium were obtained to the solutions prepared by dissolving steel samples in acid mixtures.
A solid-phase extraction method has been developed for the selective separation of nickel and applied to the determination of nickel in lead free-cutting steel. An appropriate amount of sample was dissolved in a mixture of hydrochloric acid, nitric acid, and hydrogen peroxide, and was finally prepared as a solution of 0.1 M tartaric acid–0.05% dimethylglyoxime (DMG)–0.05% Triton X-100 (pH 10). The sample solution was then passed through a column packed with polystyrene divinylbenzene copolymer resin to collect nickel as DMG complexes. Nickel complexes adsorbed on the column was eluted with 2 M hydrochloric acid and the effluent was evaporated to dryness. The residue was dissolved in 0.1 M hydrochloric acid and diluted to an appropriate volume with the same acid, followed by the determination of nickel by flame atomic absorption spectrometry. The present method was applied to the determination of nickel in a certified reference material JSS 519-1 (lead free-cutting steel, Japan Iron and Steel Federation) and 558±6 μg g−1 (n=3) was obtained (certified value: 560 μg g−1).
A unique preconcentration technique for anionic species referred to “ion-exchange preconcentration followed by ion-pair elution” was applied to separation and preconcentration of trace amounts of bismuth, lead, and antimony in digested solutions of iron and steel. First, these elements in the digested solutions were derived to anionic iodo complexes with iodide ions under an acidic condition prior to preconcentration. The derived anionic complexes were adsorbed on the anion exchanger made of cotton and trioctylmethylammonium chloride (referred to capriquat), followed by being recovered as ion-pairs of capriquat by elution with methanol. On the other hand, iron matrices did not adsorb on the anion exchanger because the iron matrices in the digested solution were reduced into divalent state with ascorbic acid not to form iodide complexes. More than 99.99% iron could be removed by rinse with a solution containing 0.8 mol dm−3 ascorbic acid and 0.5 mol dm−3 hydrochloric acid. Subsequently, the concentrates were analyzed by graphite furnace atomic absorption spectrometry. Preconcentration with 5-fold enrichment allowed us to determine trace amounts of bismuth, lead, and antimony in a digested solution of iron with detection limits (3σ) of 4.4×10−9 mol dm−3, 5.3×10−9 mol dm−3, and 3.0×10−8 mol dm−3, respectively.
To an iron sample solution were added ammonium thiocyanate, ethylenediaminetetraacetic acid, and polyoxyethylene-4-nonylphenoxy ether (nonionic surfactant, average number of ethylene oxides 7.5). The nonionic surfactant was segregated from the solution by heating at 80°C. The thiocyanato complex of iron(III) was nearly completely collected in the surfactant phase, leaving trace elements [e.g., Mn(II), Ni(II), and Bi(III)] in the aqueous phase. The trace elements were determined by graphite-furnace atomic absorption spectrometry. The proposed method was successfully applied to the analysis of high-purity iron.
The size of the blank value is enumerated in one of the factor that influence the detection limit in the fluorescence determination method. The blank value in this case is brought about by excess reagent. Therefore, the some methods for removing the excess reagent were studied under the conditions of the stability of boron-complex. To remove the influence of excess reagent, the solvent–extraction method etc. is used beside a method of changing the pH. This study attempted a fluorescence determination of boron by flow injection analysis using the methanol as a carrier liquid. As a result, the amount of reagent decreased down to 1/15 compared to that of conventional method, because of decreasing the amount of water in the sample solution. The water in the sample solution suppressed the reaction between boron and 2,3-dihydroxynaphthalene (2,3-DHN). Therefore, the decrease of water tolerates the low concentration of 2,3-DHN. Under the optimum conditions, a calibration curve was drawn, which was linear over the range of 0 to 120 ppb. The relative standard deviation at 20 ppb B was 0.9% (n=6). For the determination of a trace amount of boron in steels, the on-line separation/preconcentration of boron from iron matrix (500 ppm) was performed by using resin (Amberlite IRA 743). The limit of B determination was 0.8 ppb (1.5 ppm in steels). The determination results for boron in standard steel materials showed good agreement with the certified values.
The chemical compositions of iron ore used for iron making are mainly hematite (Fe2O3), magnetite (Fe3O4) and goethite (FeOOH). The uranium (U) and thorium (Th) of natural radioactive elements are contained with ppm level as impurity elements in these iron ore. The purpose of this study is a estimation U and Th isotopes behavior for in various iron ore by an α spectrometry. Samples were the iron ore from Australia, Brazil and India. After those samples were dissolved by a microwave method, U and Th were separated from matrix elements by the chemical processes were mainly a Fe(OH)3 coprecipitation method, a NdF3 coprecipitation method and a Nd(OH)3 coprecipitation method. Thereafter, U and Th were only separated by two steps anion–exchange chromatography. Further, Sm(III) carrier as an internal standard was added in U fraction or Th fraction, and a SmF3 coprecipitation was done with HF solution. These precipitation samples were measured by an α spectrometer system. The α peak of 238U, 234U, 232Th, 230Th, 228Th and 147Sm of these fraction were determined. As the results, U and Th isotopes in all of the iron ore were able to determine within 15% of RSD. And those were found that a relation of a radioactivity of 238U–234U was a radioactive equilibrium, and that a relation of a radioactivity of 238U–230Th was a radioactive disequilibrium. In addition, it was indicated that a positive association between their disequilibrium and the chemical compositions of ferrioxide in various iron ore.
On-site analysis of carbon in steels by laser induced breakdown spectroscopy (LIBS) was studied for the purpose of shortening of the analysis time. The LIBS system that consisted of double pulse Nd:YAG laser, an echell spectroscope and an ICCD detector was constructed. The highest emission intensity of carbon 193.09 nm line was obtained in the double pulse irradiation of the laser at wavelength 1064 nm with pulse energy of 100 mJ and 2 μs delay time of double pulses. The relative standard deviations obtained by repetition analysis of standard steel samples are about 5% in the carbon concentration range of more than 0.3 wt%. The grinding process omission was examined as a simplification of the sample pretreatment. The analytical results of carbon in steel by LIBS were compared to that by Spark-OES using samples taken at steel works for process control. It is confirmed that the analytical precision by LIBS without grinding is equivalent to that by Spark-OES with grinding. It is concluded that the LIBS technique can be applied to on-site analysis in steel works.
Nanograms of Cd are measured with a portable total reflection X-ray fluorescence spectrometer. A relationship between a detection limit for Cd and counting time is investigated. A Cd detection limit is improved with the increase in counting time, and a detection limit of 1 ng is achieved when a measurement is performed for 1800 s. Cadmium quantitative performance is shown. Yttrium is used as an internal standard element, and the Cd Kα/Y Kα intensity ratio is linear with mass of Cd in the range from 10 to 1010 ng. Detection limits for Cd obtained with or without a 30 μm thick Mo absorber are compared. The spectral background is reduced using the Mo absorber, leading to an increase in the signal to background ratio of the Cd Kα line. However, the use of the Mo absorber results in a decrease in the net intensity of the Cd Kα line. Therefore, a detection limit obtained with the Mo absorber is as low as that obtained without an absorber.
A highly sensitive method is described for the determination of bismuth by differential pulse anodic stripping voltammetry after preliminary separation of iron matrix. Most of the iron in an acidic sample solution was removed by the 4-methyl-2-pentanone (MIBK) extraction method, and MIBK dissolving in the aqueous phase was then removed by extracting three times with cyclohexane. Bismuth(III) remaining in the aqueous phase was electrodeposited on a rotating glassy carbon disk electrode at −0.6 V vs. Ag/AgCl for 600 s with stirring; the deposit was then anodically stripped at a scan rate of 50 mV s−1 to 0.2 V vs. Ag/AgCl. The calibration (peak height vs. Bi(III) concentration) graph was linear over the concentration range of 5 to 75 ng mL−1(correlation coefficient >0.999), with relative standard deviation of ca. 5% for 50 ng mL−1 (n=3). The interference from copper(II) weakened somewhat by adding 1,2-diaminopropane-N,N,N′,N′-tetraacetic acid. The minimum limit of determination of the proposed method was ca. 0.1 μg g−1, and bismuth at the μg g−1 level in iron and steel was determined with good precision and accuracy.
A simple pretreating method which consists of solid phase extraction using cation exchange extraction disk was tried for the determination of trace elements in iron and steel samples by means of inductively coupled plasma mass spectrometry (ICP-MS). The sample of 0.100 g was dissolved by nitric acid, hydrofluoric acid and hydrogen peroxide. The sample solution was adjusted by the dilution with water to pH 2.3 or more and poured into the extraction disk. The target elements retained in the extraction disk were then eluted using 10 cm3 of 3 kmol/m3 nitric acid. Quantities of the obtained target elements were determined using ICP-MS. Highly sensitive quantification was established for 8 trace elements in iron and steel, and for Al, Ca, Mn, Mg, Ba, Cd, Zn and Sr with the following detection limit [3σ; ng/g (ppb)]: Al: 0.48, Ca: 8.54, Mn: 0.09, Mg: 0.10, Ba: 0.13, Cd: 0.11, Zn: 0.21 and Sr: 0.07. This method is extremely easy, is rapid, the amount of the reagent used is a little, and a skill free and the zero emission are achieved.
Analysis of aluminum (Al) alloys was carried out using helium and argon/glow discharge mass spectrometry (He/GD-MS, and Ar/GD-MS). The relative sensitivity factors RSFX, Al over the Al matrix used for GD-MS quantitative analysis were determined using six commercial Al alloy standard reference materials. The analytes comprise 15 elements containing in Al alloys: beryllium (Be), magnesium (Mg), silicon (Si), titanium (Ti), vanadium (V), chromium (Cr), manganese (Mn), iron (Fe), nickel (Ni), copper (Cu), zinc (Zn), strontium(Sr), tin (Sn), lead (Pb), and bismuth (Bi). A glow discharge cell for disk samples was used. A 12-mm internal diameter sample mask made from Ta was employed as the part of the anode. The discharge current was 3-mA for both of discharge gas. Large dispersion was observed in several obtained RSF values, resulting from insufficient validity of certified values of some elements such as Be. Nevertheless, the dispersion in RSF values was 8% or less as RSD for most of the analytes. The Ar/glow discharge was suitable for analysis of Al alloys because the RSF-values almost were better than those obtained by measurement using the He/glow discharge. Values obtained by GD-MS for a practical Al alloy A356 well agreed with the chemical analysis result, from alloying elements like Si, minor components such as Mg and Ti, to trace elements of parts per million level. Moreover, regarding the accuracy of GD-MS analysis, minor components such as Mg, Ti, and Fe can be determined within about 2% accuracy as RSD, and trace elements below a single mass ppm within 5%, except for Cr (RSD: 24.4%), Ni (9.8%), and Sn (13.3%). We established an eco-friendly analytical method without acid-decomposition by chemical reagents.