ABSTRACT

Radioisotope concentrations, such as ²³⁸U and ²³²Th, of bulk sediments are good environmental indicators and useful for radiometric age determinations. Pressed pellets of sediment can be analyzed by laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) with little preparation time. However, the matrix effects caused by the different properties between the external standard and measured sample may yield unreliable results. This study examines the preparation method of pressed powder pellets using geological rock reference materials, and tests them as external standards. Powders of the rock reference materials (JA-1, JA-2, JB-1 and JB-2 from the Geological Survey of Japan), a lacustrine sediment (4A-18 from Lake Baikal, Russia), and solid glass reference standards (SRM 610 and 612 from the National Institute of Standard and Technology) were analyzed. In calculating ²³⁸U or ²³²Th concentrations, one sample was selected as an external standard and other six samples were treated as unknown samples. ²⁹Si was chosen as an internal standard. When the grain size is small enough (about <10 μm) compared to the laser spot size, and powders are sufficiently homogenized, relative standard deviation (rsd) of analytical data ²³⁸U was small enough (average: 14%) to produce reliable data. In the case of ²³²Th, accuracy was affected by the matrix effect regardless of grain size: the concentrations of pressed pellets measured by SRM glasses as external standards are always lower than the reference values, while those measured by other pressed pellets as external standards are almost concordant to the reference value within a range of ±10.1% RSD. This is most likely due to incomplete digestion of large particles from the pressed pellet in the ICP-MS. Therefore, the pellet-pellet combination for an unknown sample and an external standard is recommended although behavior of different matrix element composition or different mineralogy should be investigated further. In contrast to ²³²Th, ²³⁸U appears to be free from matrix effect so that both pellet-pellet and pellet-glass analyses are feasible.