LIBS Info: Element Analysis

Title Authors Material Detector Spectrometer Software
Multi-element analysis of iron ore pellets by Laser-induced Breakdown Spectroscopy and Principal Components Regression D.L. Death, A.P. Cunningham, L.J. Pollard Iron Ore Princeton Instruments OMA SPEX 1877C GNU Octave
Laser: Excimer
308.0000nm
NonemJ
50.000Hz
Gate Delay: 2.000us
Gate Width: 4.000us
PCR is used to determine Fe, Al, Si, Mn and K in Iron Ore pellets. Laser beam was slot shaped rather than the usual round. Samples were ring milled, the pressed into pellets for analysis. The original 20 samples were skewed towards higher Fe concentrations, so an additional 11 samples produced by mixing original samples were made to give a more even concentration spread. Dark noise was subtracted, then a normalisation to regions of the spectra that showed little variation (RSD < threshold across calibration set). 16 samples were used in calibration, with the remaining 15 used as validation. A PCR model for P was tested, but the results were not good.
Element Detection Limit (ppm) Wavelength (nm) Other Wavelengths (nm) Calibration Method Calibration Range (ppm) Notes


Element RMSE (ppm) Wavelength (nm) Calibration Method Notes
Fe 0.4900 % -10.0000 PCR Calibration using 250nm region
Fe 0.6400 % -10.0000 PCR Calibration using peaks around the 400nm region
Al 0.5300 % -10.0000 PCR Calibration using peaks around the 250nm region
Al 0.8700 % -10.0000 PCR Calibration using peaks in the 400nm region
Si 0.1700 % -10.0000 PCR Calibration using peaks in the 250nm region
Si 0.2000 % -10.0000 PCR Calibration using peaks in the 400nm region. This is surprisingly close good as Silicon has only a couple of weak peaks in this region.
Mn 0.0170 % -10.0000 PCR Calibration using peaks in the 400nm region. This model is not as good as Fe, Al or Si for these samples (R^2 = 0.84)
K 0.0060 % -10.0000 PCR Calibration using peaks in the 765nm region