A modified spectral standardization method for uranium ores measurement using laser induced breakdown spectroscopy

The high signal uncertainty is the key factor hindering the quantification performance of laser induced breakdown spectroscopy (LIBS) technology. Different plasma parameter compensation algorithms for reducing signal uncertainty have been proposed. Among them, spectral standardization is a representative, which can reduce signal uncertainty by compensating for the variations of plasma parameters, especially using multiple independent spectral lines of the measured element to compensate for the fluctuation of total number density. However, it's hard to find enough independent lines for cases such as uranium in ores measurement. This work proposes an updated version to reduce the signal uncertainty for these cases.

On the basis of spectral standardization method, the new method uses another element information of the matrix to compensate the fluctuation of total number density for the measured element. Specifically, multiple spectral lines of another element are first used to characterize its total number density, and then the total number density of the measured element is obtained through the correlation between the two elements. Finally, the new method is evaluated on uranium ores. Results showed that after using the information of the silicon element to estimate the total number density of uranium, the relative standard deviation(RSD) of the U II 409.01 nm was reduced from the original 23.03 %% to 9.41 %, the quantitative model determination coefficient (R_C²) of uranium was increased from the original 0.9129 to 0.9921, and the root mean square error of prediction (RMSE_P) was reduced from the original 0.2586 to 0.0670.

The modified spectral standardization method, which using another element information to estimate the total number density of the measured element, can greatly reduce the signal uncertainty and improve the quantitative analysis performance of uranium in ores. What's more, it provides a new signal uncertainty reduction method for elements without enough independent spectral lines, such as the spectral with severe interference, which expands the application scope of spectral standardization method.