BACKGROUNDInductively coupled plasma-mass spectrometry (ICP-MS) has been widely used in the determination of trace elements in carbonate rocks. Due to the low Ni (1.6-50.5μg/g) and Sc (0.3-6μg/g) in carbonate rocks, the signal values are obviously interfered by high CaO (up to 56%) and MgO (up to 21%) during ICP-MS determination, resulting in the test value much higher than the true value.

OBJECTIVESTo solve the problem of non-mass spectrum interference and mass spectrum interference of Ni and Sc during ICP-MS analysis of carbonate rocks and use appropriate correction method.

METHODSThe single standard series of Ca and Mg and national first-level reference materials of carbonate rocks were used to study the interference of high content of Ca and Mg on Ni and Sc in the carbonate during ICP-MS analysis. Testing of the single standard series was aim to explore ways of interference on Ni and Sc by high content of Ca and Mg in solution. The national first-level reference material of carbonate rock was further selected as the calibration carrier to eliminate the matrix effect of Mg. At the same time, according to the good linear relationship between the content of CaO in the sample solution and the interference degree of Ni and Sc, the interference equations of CaO and △Ni and △Sc were fitted respectively, and used for interference deduction of Ni and Sc in several national first-level reference materials and unknown samples of carbonate rocks. The accurate test values of Ni and Sc in carbonate rocks by ICP-MS were obtained.

RESULTSIt was found that the high content of Mg had a non-mass spectrometric interference matrix effect on the analysis of Ni and Sc. High content of Ca forms oxides, hydroxides and polyatomic ions, resulting in mass spectrometric interference on Ni and Sc. The degree of interference had a good linear relationship with the Ca content in the solution. Compared with a single standard to perform interference correction on actual samples, this method used national first-level standard materials as the calibration carrier, which overcomed the interference of matrix effects. Verified by GBW07108 and other five national primary standard materials of carbonate rocks, test values agreed with the standard values, with a relative standard deviation (RSD, n=10) of less than 5.5%. Correction results of the unknown carbonate samples were compared with the results of the inductively coupled plasma-optical emission spectroscopy (ICP-OES) and the X-ray fluorescence spectrometer (XRF), respectively, the relative deviation was less than 15%.

CONCLUSIONSThe correction method proposed in this paper has solved the mass spectral interference and non-mass spectral interference in ICP-MS analysis of Ni and Sc in carbonate rocks. The method is simple and feasible, and the results are accurate and reliable.