Graphite furnace for alternative combination with d.c. arc or inductively coupled plasma. Introduction and analysis of solid samples

Summary A graphite furnace and d.c. arc combined source with the use of halogenating atmosphere in the furnace is described. The construction makes possible the introduction and analysis of both liquid and solid samples at minimum risk of contamination. Spectrographic detection resulted in background equivalent mass values below 10 ng for 22 from 36 elements applied in solution (Ar + CCl₄ atmosphere and NaCl matrix). The intensity-time profiles indicated high rates of distillation of the impurities (including B, Cr, Hf, Mo, Ti, V, W and Zr) from silicon carbide and silicon nitride samples in the first 40 s of heating under chlorination (CCl₄ vapour). Two versions of the exit part of the furnace are described for coupling to an ICP source; one of them has been used for the determination of titanium in alumina (spectrometric detection). Standardization with solution standards for solid sample analysis was unsuccessful. Based on the analytical curve obtained with variable amounts of solid standards (5–20 mg), a detection limit of 0.32 g/g Ti was found for 20 mg alumina.     

电感耦合等离子体原子发射光谱(ICP-AES)法测定钢中硼时样品溶解损失及空白控制研究

用电感耦合等离子体原子发射光谱(ICP-AES)法研究了钢中硼的测定。实验发现,分析线在硼182.64nm处无硼铁基体空白值低于纯试剂(无铁)空白值。样品分析时,应进行无硼铁基空白实验,结果计算只能用此空白值。对普通玻璃烧杯和聚四氟乙烯烧杯的酸溶样品测定结果进行了比较,结果表明,测定结果和空白值相近且都很低,无硼普通玻璃烧杯也能满足钢中低含量硼的测定需要。通过对实验与报道结果比较分析表明,硝酸或王水加热溶解样品得到的溶液中硼没有损失,此溶液可测定钢中酸溶硼;用高氯酸发烟处理后,样品中酸不溶硼可能会溶解,但硼会挥发损失。因此,高氯酸发烟处理不能测定硼。     

Determination of boron in silicon by SIMS with the matrix ion species ratio method

Summary In quantitative SIMS, the oxygen content of the sample surface proves to be a very important analytical parameter. The matrix ion species ratio method (MISR) was used to investigate the influence of the presence of oxygen on the determination of boron in silicon. By analysing standard samples at conditions of different oxygen coverage, the relation between the sensitivity of boron and the oxygen content of the sample surface, as indexed by the SiO⁺/Si³⁺ matrix ion species ratio, was established. The influence of the primary ion current density on this relation was investigated. The MISR procedure proves to be a suitable method to perform quantitative determinations (relative error < 10%) in conditions of changing oxygen content on the sample surface.

---

Bestimmung von Bor in Silicium durch SIMS mit der Matrix Ion Species Ratio Method

     

Analysis of silicon carbide powders with ICP-MS subsequent to sample dissolution without and with matrix removal

ICP-MS has been employed for the analysis of silicon carbide powders in connection with high pressure acid decomposition without and with matrix removal by evaporation. The powder is decomposed by treatment of a 250 mg sample with a mixture of HNO₃, H₂SO₄ and HF. Prior to the analyses with ICP-MS the solutions have to be diluted to a matrix concentration of 500 g/ml related to SiC in order to realize full long-term precision. The results obtained for Li, B, Na, Mg, Al, Ca, Sc, Ti, V, Cr, Mn, Fe, Ni, Co, Cu, Zn, Ga, Sr, Y, Zr, Nb, Ag, Cd, In, Sn, Sb, Ba, La, Ce, Pr, Nd, Hf, Ta, W, Tl, Pb, Bi, Th and U in SiC powder S-933 are shown to be in good agreement with those of independent methods, such as INAA, ICP-AES with slurry atomization and ICP-AES subsequent to sample decomposition. For extending the use of ICP-MS to elements such as Mg, Ca, Sc and Ti at the relevant concentrations in SiC powders, a more effective matrix removal by evaporation of the decomposition solution to near dryness has been successfully applied. Its advantages have been proven by the results of high resolution ICP-MS. It has been found by analyses of the treated sample solutions for the residual Si and C with ICP-MS that over 99% of the matrix and also of the acids used for decomposition are removed. For B, Al and Fe losses were found to occur at concentration levels of some g/g, 200 g/g and 300 g/g, respectively, and all other elements were detected with very good recoveries. For all 36 elements investigated in this work the detection limits could be improved from the ng/g to the pg/g range by removal of the matrix. The analytical range could be improved, in particular for In, Tl, Bi and U.Dedicated to Professor Dr. Dieter Klockow on the occasion of his 60th birthday     

Rapid and Simple Method for the Determination of Vanadium in Beer by ICP AES with Ultrasonic Nebulization

An accurate and reproducible method for the direct determination of vanadium in beer using inductively coupled plasma atomic emission spectrometry (ICP AES) with an ultrasonic nebulization system is described. The analytical conditions were optimized in order to achieve the appropriate sensitivity. Samples were digested with a 1 : 1 nitric acid–perchloric mixture. The analytical detection limit was 0.4 g/L, and the relative standard deviation was 2.8% at the 10 ng/mL V levels. The method was applied to determine vanadium (15.9–244.3 g/L) in 11 different samples of beer widely consumed in the Argentine Republic.     

Trace analysis of selected benzidine and diaminodiphenylmethane derivates in urine by means of liquid chromatography using precolumn sample preconcentration,UV and electrochemical detection

Summary For the title compounds 4,4-DADPM, MOCA, 3,3-DCB, 4-ADP and 4-ADPA listing on the EPA priority pollutant list, an analytical practicable, reliable, reproducible and sensitive procedure is required. Therefore a new method has been developed for the routine determination of these toxic aromatic amines in urine at the ppb level. The quantitative determination of amines is a suitable procedure of occupationally exposed persons. Urine sample preparation is done using simple liquid-liquid extraction followed by a precolumn enrichment (PRP₁-material; Hamilton). Breakthrough measurements were done using an enrichment column packed with PRP₁ material. The capacities of the studied amines ranged from 21.9mg/g to 96.6mg/g, while influent concentrations differed from 28.3mg/l to 332.0mg/l. The advantages of electrochemical detection regarding to selectivity and sensitivity are clearly indicated in this paper. Separation has been achieved applying reversed-phase-high-performance-liquid chromatography (LiChrosorb RP 18/5m) followed by electrochemical or UV-detection. The detection limits employing an electrochemical detector at a potential of 1 V range from 2.2ng to 12.1ng. UV detection at 254 nm and 280 nm is about 10–100 times less sensitive. Recoveries from spiked water samples at the 5ppb levels were 75% to 96% respectively. The standard deviation of the developed procedure varies from 5.3% to 14%. Day-to-day repeatability is good.     

Investigation for the determination of ultra trace amounts of Co2+ complexed with 1,10-phenanthroline by adsorptive voltammetry

Summary The spontaneous adsorption of the Co²⁺ complex with 1,10-phenanthroline on HMDE was utilized for the development of a voltammetric analytical method employing phase selective alternating current, with first harmonic AC₁ in phase, =0°, voltammetry. The dependence of the analytical voltammetric response on physical and chemical parameters in various supporting electrolytes was investigated in detail. Optimum conditions for Co²⁺ determination were established. The method is very sensitive and selective. The determination limit is 205 ng/l Co²⁺ (adsorption time, t_ads=20 min). The precision and accuracy were evaluated: 1.84 g/l Co²⁺, RSD=3.24%, RE=–0.54%; for 205 ng/l Co²⁺, RSD=4.85%, RE=–3.90%. The results showed good reproducibility and reliabiality. The method was applied to natural waters, wines, mussel tissue and aquatic algae certified reference material.     

Trace analysis of chlorine in high-purity refractory metals with isotope dilution mass spectrometry

Summary A method for the determination of chlorine in high-purity molybdenum and tungsten with isotope dilution mass spectrometry (IDMS) has been developed using a ³⁷Cl enriched spike solution. After dissolution of the sample in H₂O₂, chlorine is separated by AgCl precipitation. Negative Cl^- thermal ions are produced in a double-filament ion source using an easy-to-handle and cost-efficient quadrupole thermal ionization mass spectrometer. The detection limit, which is mainly influenced by contaminations from the laboratory air, lies between 0.006 and 0.1 g/g depending on the variation of the blank values. Chlorine concentrations between 0.07 and 10.5 g/g are analysed in different tungsten and molybdenum samples. A comparison with other sensitive analytical methods shows the urgent necessity of developing alternative methods for the determination of chlorine traces in refractory metals.Dedicated to Professor Dr. V. Krivan on the occasion of his 60th birthday     

Determination of trace impurities in electronic grade quartz: Comparison of inductively coupled plasma mass spectroscopy with other analytical techniques

An analytical procedure for the determination of uranium and thorium in the sub-ng/g range as well as of other trace elements in the ng/g to g/g range in high purity quartz samples is described. The results obtained by inductively coupled plasma mass spectroscopy (ICP-MS) are compared to those obtained by other analytical techniques (instrumental neutron activation analysis, INAA; flame atomic absorption spectrometry, AAS; Zeeman graphite furnace atomic absorption spectrometry, ZGFAAS; total reflection X-ray fluorescence analysis, TRFA; direct current arc optical emission spectrometry, DC-arc OES; and X-ray fluorescence analysis, XRFA). For the ICP-MS measurements, the decomposition of the samples is carried out with HF/HNO₃/H₂SO₄-mixtures. The results obtained by the different methods show reasonable agreement. For uranium and thorium, ICP-MS proves to be the most sensitive method: detection limits of about 50 pg/g can be achieved for both elements.Presented in part at the 1989 European Winter Conference on Plasma Spectrochemistry, Reutte, Austria     

Lead determinations in human bone by particle induced X-ray emission (PIXE) and graphite furnace atomic absorption spectrometry (GFAAS)

Chronic lead (Pb) intoxication has been linked to Alzheimer's disease (AD). Lead, like many heavy elements, tends to accumulate in bone. PIXE is a powerful analytical tool which permits the determination of Pb at the g/g level without requiring sample digestion. GFAAS is one of the most sensitive methods for the determination of Pb and is capable of determining ng/g levels in solution. For bone analyses by GFAAS, sample dissolution and a matrix modifier are required. Rib bone samples were analyzed for Pb by PIXE and GFAAS. IAEA Animal Bone (H-5) was used as a secondary standard for Pb with both methods to ensure accuracy. The range of Pb concentrations in human rib bone was 1.4–11.5 g/g for the trabecular surface by PIXE, 1.3–45 g/g for the cortical surface by PIXE, and 1.54–11.75 g/g for whole bone by GFAAS. No significant difference (p>0.05) was found for AD versus control for either surface or for whole bone.     

A radioreagent method for determination of traces of phosphorus in high-purity silicon

The present study explores the feasibility of the determination of phosphorus at the extreme trace levels in high-purity silicon by radioreagent method. After silicon dissolution with hydrofluoric and nitric acids and matrix volatilization, 12-molybdophosphoric acid (12-MPA) is formed by the addition of the radioreagent,⁹⁹MoO ₄ ^2– , in nitric acid medium and then extracted into isobutyl acetate. By plotting the phosphorus content against the radioactivity of⁹⁹Mo in the organic phase, a linear relationship persisting down to 5 ng is obtained. Special effort has been made to the elimination of the unreacted⁹⁹MoO ₄ ^2– reagent and the optimal control of phosphorus blank introduced through the multistage analytical procedure in order to ensure reliable determination of phosphorus at the ppb level.     

Determination of ultratrace boron concentrations in uranium oxide by isotope dilution-thermal ionization mass spectrometry using a simplified separation procedure

Boron was determined at sub-ppm levels in uranium oxide by ID-TIMS. Following nitric acid dissolution, a solvent extraction method was adapted for extraction of boron by 5% 2-ethylhexane-1,3-diol in chloroform followed by evaporation of the extractant in presence of sodium carbonate and mannitol. This procedure allowed TIMS analysis of boron by monitoring the Na₂BO ₂ ⁺ ions formed from the sodium borate salt in the ion source. Reproducibility of about 1% was obtained for boron determination in the three different uranium oxide samples containing sub-ppm amounts of boron.     

Trace analysis of ceramic surfaces by laser ionization mass spectrometry

Summary Laser ionization mass spectrometry (LIMS) is an analytical method for the simultaneous determination of concentrations of trace elements in solid samples and for the analysis of layers with thicknesses >1 m. This laser-induced surface analytical method is limited by the laser focus and crater depth of the laser system applied. Results of mass spectrometric trace analysis on a silicon carbide surface, ZrO₂ and high-T_c superconducting ceramics are discussed.     

Atomic emission and atomic absorption spectrometric analysis of high-purity powders for the production of ceramics

Summary Direct analysis methods and multistage combined analytical procedures for the determination of impurities at the g/g level and the upper ng/g level in high-purity powders of Al₂O₃, AlN, Si₃N₄ and SiC are described. Results obtained with a novel direct slurry-atomization technique using a Babington nebulizer and inductively coupled plasma optical emission spectrometry (ICP-OES) are presented. A comparison of analysis results of combined analytical procedures including wet chemical decomposition and determinations with graphite furnace atomic absorption spectrometry (ETAAS) or ICP-OES with those of slurry-atomization ICP-OES show the capabilities of this technique for routine analysis in production control. Detection limits for Al, B, Ca, Co, Cu, Fe, Mg, Mn, Si, Ti, W, V, and Zn in the matrices mentioned are between 0.03 and 2.5 g/g. For elemental concentrations 10 g/g relative standard deviations of the measurements are generally below 10%. The technique is shown to be a powerful tool for trace determinations in powder samples. This is shown by its use for analysis of a series of the ceramic powders mentioned and comparative results of other direct techniques such as total reflection X-ray fluorescence spectrometry and instrumental neutron activation analysis.

---

OES und AAS von hochreinen Pulvern für die Keramikherstellung

---

Herrn Prof. Dr. G. Tölg zum 60. Geburtstag von seinen Mitarbeitern gewidmet     

Contribution to the clarifying of the action of thermochemical additives on the modification of a matrix silicon carbide using scanning electron microscopy and energy dispersive X-ray analysis

Summary Previous investigations have shown that the total evaporation of impurities from silicon carbide powder can be achieved by its thermochemical transformation to BaSiO₃, BaF₂ and CoSi (CoSi₂) under the action of the modifier CoF₂ + BaO, through in situ processing in an electrothermally heated graphite crucible [13]. The aim of the present work has been to describe a methodology to perform a rational selection of various potential candidates for modifiers, to make successful combinations between various chemical additives, to understand the reason(s) for an effective/ineffective action of a particular additive, i.e. to find the optimum conditions for each type of ceramic matrix chemical modifier. A combination of scanning electron microscopy (SEM), energy dispersive X-ray analyses (EDX) and X-ray diffraction analyses was used to confirm the identification of the alterations to the silicon carbide matrix. Indirectly, the present work supports the view that the aim justifies the means particularly when the aim is to reach a complete evaporation of the impurities from a carbide matrix, which as is well known, is not an ordinary or easy task.     

Inductively-coupled plasma atomic emission spectroscopic determination of trace impurities in ZrO2-powder

Four independent procedures including one using slurry nebulization ICP-AES were developed for the trace analysis of ZrO₂ powders. They were evaluated with respect to detection limits, blank values, interferences, accuracy and precision. For the procedures I–III ZrO₂ powder was decomposed by fusion with a 10-fold excess of NH₄HSO₄ and subsequent dissolution of the melt in either water or, after evaporation of NH₄HSO₄, in diluted HNO₃. In procedure I the solution was directly analyzed by ICP-AES, which was optimized with the aid of a simplex algorithm. In procedure II Zr was separated by extraction from 6 mol/l HNO₃ with a 0.5 mol/l solution of 2-thenoyltrifluoroacetone (TTA) in xylene. More than 99.5% of the Zr was removed and more than 95% of the trace elements retained. In procedure III the matrix was separated by its precipitation as ZrOCl₂·8 H₂O from a (1:4) HCl-acetone medium. More than 98% of Zr were removed and more than 90% of the trace elements were retained. In procedure IV the ZrO₂ powder was dispersed by ultrasonic treatment in water acidified with HCl (pH 2) and the slurry was directly analyzed by ICP-AES using a Babington nebulizer. The optimization and the analytical features of this procedure will be described in a subsequent paper. In all procedures the calibration was performed by standard addition and matrix matching was not necessary. The detection limits varied from 0.3 g/g (Ca) to 10 g/g (Al). The standard deviations obtained were 1–10% depending on the element and its concentration in the sample. The results of the procedures for 6 commercially available fine ZrO₂ powders were found to agree for Al, Ca, Fe, Mg, Na, Ti and Y. A good agreement between the results of the procedures using matrix separation was also observed for Cu, Mn, V, but the concentrations of these elements found by methods without matrix separation were considerably higher. Except for Ca and Mg the blank values encountered were below the detection limits.On leave from Department of Analytical Chemistry, Technical University, PL-00-664 Warsaw, Poland     

Determination of Silicon Impurities in High-Purity Arsenic Using Atomic Emission Spectrometry with Matrix Distillation from a Tipped Electrode

A procedure for analyzing high-purity arsenic by atomic emission spectrometry (AES) was proposed that provides the preconcentration of silicon and other nonvolatile impurities in a crater of a graphite electrode using matrix distillation from 2-g sample portions as As₂O₃. The procedure is characterized by a low correction for the blank experiment. The detection limit for silicon was 4 × 10^–7 wt %.     

Chromatographic Determination of Silicon and Phosphorus as Molybdic Heteropoly Acids with Preconcentration

The chromatographic behavior of molybdic heteropoly acids of silicon and phosphorus as ion pairs with tetrabutylammonium bromide was studied by ion-pair high-performance liquid chromatography on the C₁₈reversed phase (UV detection at 310 nm). Heteropoly acids were preconcentrated as ion pairs with tetrabutylammonium bromide to increase the sensitivity of the determination. Optimal conditions were selected for the chromatographic determination of silicon and phosphorus in water in the presence of each other. Detection limits for silicon and phosphorus are (1.1 × 0.3) × 10^–3and (6.7 × 1.2) × 10^–3g/mL, respectively. Calibration plots are linear in the concentration ranges 0.01–0.1 g/mL (silicon) and 0.02–0.15 g/mL (phosphorus). The procedure was used for the analysis of distilled water.     

On the use of relative deceleration values for the determination of silicon by PIGE

By spiking the samples with a compound containing both a non-analyte and an analyte element, to which a relative deceleration property for 5 Me V protons has been ascribed, relative deceleration values for the samples could be obtained by measuring the prompt -yields induced in the non-analyte element. These values are used to correct for matrix effects in elemental analysis using PIGE techniques. In addition to this, the analytical results thus obtained can also be compared to those obtained by the simultaneous application of the analyte spiking technique to the collected yield data. These non-analyte spiking approaches were used to analyze silicon in polymer materials. These methods were tested by analyzing the two reference standards BCS 308 and Standard Chrome Ore XXXI. Li₂SiO₃, containing both the non-analyte as well as the analyte element, was selected as comparator and spike. Results obtained for the polymer samples were also compared to those obtained by PIXE.