Some results of fundamental studies on gas-phase sample introduction methods in analytical atomic spectroscopy, i.e., atomic absorption, atomic emission and atomic fluorescence spectrometry, are reviewed on the basis of a series of researches carried out in the author's laboratory. The gas-phase sample introduction techniques can be divided into three groups: i) hydride generation, ii) cold-vapor generation of mercury and iii) analyte volatilization reactions. The analytical figures of merit of the gas-phase sample introduction methods have been described in detail. Hydride generation method has been coupled with atomic absorption spectrometry of arsenic, bismuth, germanium, lead, antimony, selenium, tin and tellurium, with inductively coupled plasma (ICP) atomic emission spectrometry of arsenic, tin, selenium, antimony and bismuth, and with fluorescence spectrometry of arsenic, antimony, bismuth, selenium, tellurium, lead and tin. Cold-vapor generation method for mercury has been combined with atmospheric pressure helium microwave induced plasma (MIP) atomic emission spectrometry and atomic fluorescence spectrometry. Furthermore, analyte vaporization reactions have been employed in ICP atomic emission spectrometry of iodine and in MIP atomic emission spectrometry of iodine and bromine. As a result, when compared with conventional solution nebulization, a great improvement in sensitivity has been attained in each instance. In addition, the present techniques coupled with analytical atomic spectrometry have been successfully applied to the determination of trace elements in a variety of practical samples.

View full abstract

Wet-cleaning processes using chemicals will continue to play an important role in VLSI manufacturing. It is, thesefore, vital to reduce metal concentrations and particles in chemicals, because the dimension of VLSI continues to shrink. The metal concentrations are believed to have an adverse effect on pn junctions leakage current. In particular, the number of residual heavy atoms on wafer surface is to be kept within 10¹⁰ atoms/cm². Metal concentration level in the chemicals is not less than 1 ppb at present. The true impact of particle control, however, has not yet been realized. The constituent elements of particles in the chemicals were analyzed by an electron microanalyzer (SEM-EDX) which can detect light-element as well. Samples used in this study were prepared by filtering various chemicals with 0.2 μm Nuclepore filter. After undergoing filtering, the filter was repeatedly put on sheets of wet filter paper until the contamination of the back side of the filter was removed. Then, the filter was dried on a filter paper under clean atmosphere. Carbon or Pt-Pd film was then sputter-deposited on the surface of the filter. From the results of analysis, it was found that the particles were mainly composed of Si and unknown materials with a little amount of Al, Mg, Fe, Cr, Ti etc. In case of H₂SO₄, only Si concentration increased a little by an elution test for 2 months.

View full abstract

One milligram of powdered quartz sample was mixed with 1 mg of pulverized graphite to determine Mn and Cu by direct atomization-AAS in a miniature graphite cup placed in graphite crucible. Ten microliters of 0.2 ppm Ni solution was added to the sample powder as a matrix modifier. Optimum atomization conditions were as follows: drying at 80120°C (30 s), pyrolyzing at 1000°C (30 s) for Mn and 800°C (30 s) for Cu, and atomization at 2500°C (15 s) for Mn and 2600°C (15 s) for Cu. Calibration curves were made by using aqueous standard solutions. Results for Mn (0.0290.19 ppm) and Cu (0.0330.19 ppm) in synthetic quartz samples and rock crystals agreed well with values obtained by dissolved samples. Relative standard deviations (n = 5) were 2.8% for 0.037 ppm Mn and 1.9% for 0.099 ppm Cu.

View full abstract

A method is described for the quantitative analysis of total polychlorinated biphenyls (PCBs) separated by capillary GC. Equal amount of KC-300, -400, -500 and -600 are mixed and dissolved in hexane to prepare the reference solution (80 mg/l). The reference solution is chromatographed by capillary GC using atomic emission detector. From the chromatogram acquired by carbon emission, the abundance ratios of biphenyl groups distributed over the chromatogram are estimated. Chlorine substitute number of individual congener is also estimated by using the suggested equation. GC/MS is used for the confirmation of chlorine substitute number. Combining these data, the percent abundance of individual congener (CB%) is estimated. Reliability of the method is tested by analyzing artificially mixed PCB samples, and good results are obtained.

View full abstract

A flow injection system with multi-wavelength detection is presented for rapid and simple simultaneous determination of iron and titanium in various aluminum alloys. The analysis is based on the complex formation of iron and titanium with diantipirylmethane in a continuous flow system and on-line monitoring of absorption spectra of those metal complexes. Concentrations of iron and titanium are calculated by solving two simul-taneous equations consisting of the absorbances measured at two chosen wavelengths of λ₁ and λ₂ and molar absorptivities of each metal complex at λ₁ and λ₂. The results of the analysis of standard aluminum alloy samples by the proposed FIA system (λ₁=388 nm, λ₂=540 mn) correspond well with the reference values for iron and titanium and showed good reproducibility with relative standard deviations less than 3%. The sample throughout was about 25 h^-1.

View full abstract

Fluoride in water samples of given fluoride concentrations has been determined based on the “JIS K 0102 (34.1)” method. The following two problems have been found: 1) near the lower limit of the concentration range in this method, an error of about +25% occurs; 2) neither the available period nor the manner of storing the lanthanum-Alizarine complexone solution is described. The following two proposals have been made for dealing with the above two problems: 1) the value of the lower limit for the determination of fluoride should be raised; 2) the complexone solution remains stable for about 40 days when the solution is stored in a brown bottle in a dark and cool place.

View full abstract

The procedure of acid digestion of biological samples was improved to minimize Al contamination. Quartz digestion flasks and the stoppers were cleaned with hot HNO₃ and they proved to be free from aluminium. Reagents (HNO₃ and H₂O₂) containing very small known amounts of Al were selected. Blank measurements were always carried out by using 3 to 6 flasks per sample, and taking the average. The blank values was reduced to 0.0110.029 μgAl per flask. By this procedure, the Al content in a NBS standard bovine liver sample (1577a, non-certified value: 2 μg/g) and in human blood was found to be 1.39±0.32 μg/g and 0.01 μg/g, respectively. Recovery of Al added to the bovine liver sample was 93113%.

View full abstract

An analytical method has been established for the determination of ruthenium in glomerular basement membrane (GBM) by graphite furnace AAS. The tissue was decomposed by 25% tetramethylammonium hydroxide. By ashing the solution (10 μl) at 1200°C for 30 s followed by atomizing at 2500°C for 5 s by using argon as a purge gas, atomic absorption of ruthenium was measured at 349.9 nm. The effect of diverse elements was suppressed by the use of peak area measurement and the optimum ashing temperature. The relative standard deviation of the proposed method was 10.3% at 56.1 ± 5.8 μg/mg GBM protein. The limit of detection was 5.5 ppb.

View full abstract

Hematite (α-Fe₂O₃) standards for local analysis were prepared by the flux growth method. Five kinds of hematite crystals containing 0.032.6 mass% Al₂O₃ were grown in the Na₂O·2B₂O₃ or Na₂O·4B₂O₃ flux by Ostwald ripening. The crystals were separated from the flux and then sintered at 1310°C for 3 d in air. The Al distribution in the sintered tablet was examined by electron probe X-ray and ion microanalyzers. In the former analysis, the RSD for the Al concentraions was found to be ca. 2% in the concentration range 0.62.6 mass% Al₂O₃. The above RSD value was somewhat beyond the normal margin of error in measurements. In the latter analysis, the RSD value for the relative intensity ratio, I(²⁷Al⁺)/I(⁵⁷Fe⁺), was found to be 46% for different parts and 14% for depth, in the concentration range 0.030.6 mass% Al₂O₃.

View full abstract

The difference of refractive index between the sample solutions injected and the carrier solution (the reagent solution) causes problems in detection if an FIA system with a spectrophotometric detection is applied to various samples containing solvents or salts. To recognize this phenomenon, a dual channel FIA apparatus was designed and distilled water was delivered in both streams. Sample solutions were injected into the stream and flow signals were recorded. Several water miscible solutes such as NaCl, urea and dextrose were selected as samples, and Methyl Orange was used as a standard colored sample. The signal intensity changed depending upon the refractive index. Therefore, the effect of the matrix in sample solution should be considered.

View full abstract