High Precision Boron Isotope Analysis by MC-ICP-MS

The precision of boron isotope analysis with MC-ICPMS depends on spectral interferences (eg. ⁴⁰Ar⁺⁺⁺⁺), instrumental drift, mass bias and memory effects. Spectral interferences and instrumental drift can be corrected using blank calibration and the external standard method, respectively. Mass bias of boron isotope analysis will not reduce the precision if the samples and standards are all ~0.1ppm, which meet our sample condition well.  Memory effects from sample inlet pipe, spray chamber and the cones can plague MC-ICPMS analyses. Especially, boron is very hard to be eliminated from the inlet system. For most isotope analyses, this memory can be reduced by extending the washing time. However, washing time of even 1 h is insufficient to reduce the memory of boron below 1% using routine solvents. In our research, we found that the mix acid solution of 2%HNO₃+0.1%HF is effective to wash out the memory of boron. The washing time to reduce the memory below 1% can be shortened to ~12 minutes. On the other hand, the researchers always neglected the boron isotope difference between the memory and the later sample. If a precision of 0.1‰ is required, the difference between sample and standard should be less than 10‰ when the memory below 1%. However, some samples have much larger boron isotope difference to the standard. Therefore, we improve the washout process designed as follows. The inlet system will be washed out with 2%HNO₃+0.1%HF until the memory below 1%, and then a new sample or standard is introduced for about 1 minutes, following  one more washing out till 1%. Finally, the same sample or standard is introduced and analyzed. Our present study indicates that the precision of boron isotope determination by MC-ICP-MS (NEPTUNE) is better than 0.08‰ (SD) to 0.1ppm boron samples and a complete analysis procedure takes about 20 minutes.