抄録
When trace elements in an ultra high-purity iron were determined, an analytical blank on a chemical analysis interfered with the determination of them. Similarly, the existence of major component in a sample interfered with the determination of trace elements. Therefore, in order to determine trace elements, it was examined that how to remove or decrease an analytical blank and how to separate trace elements from major component of a sample. For example, when C and S in an ultra high-purity iron were determined by combustion/infrared absorption method, an accelerator for combustion of an analytical sample contained C and S as blank. The accelerator consisted of a mixture of tungsten and tin. By heating it in ambient atmosphere, carbon blank was removed to be zero and sulfur blank was decreased to lower level. Consequently, when detection limit was defined as a value corresponding to 3 times of the standard deviation of blank value, it of carbon was infinitesimal and it of sulfur was 0.2 \\microgram g−1. Infinitesimal detection limit of carbon resulted from zero of the blank value of carbon. For that reason, detection limit of carbon was read as 0.01 \\microgram g−1 of carbon that was minimum scale for carbon on the analytical instrument, because the standard deviation of the blank value could not to be calculated. Independently, in order to determine trace Sn, Ag and Au in a high-purity iron, these elements were separated from major component (iron) by co-precipitation. Metallic Pd was used as co-precipitant. Metallic Pd had been rarely used as co-precipitant. The elements were determined by electrothermal atomic absorption spectrometry (ET-AAS). When these elements were detected by ET-AAS, the Pd was available as chemical modifier. By the separation, content 0.002–0.02 \\microgram g−1 of the elements was determined.
