Study on the spectral characteristics and analytical performance of pulverized coal using laser-induced breakdown spectroscopy under a fast physical constraint
The analysis of coal quality provides strong support for efficient coal utilization and environmental protection. However, finding a simple and fast sample preparation method to get good spectral quality is the main challenge for achieving online analysis through laser-induced breakdown spectroscopy (LIBS). In this work, a sample preparation device based on a physical constraint was designed to prepare pulverized coal quickly. For the process of sample preparation and detection, pulverized coal was confined in a closed cavity in the form of coal pillars; plasma was also confined in a narrow opening on the sidewall of the device. Therefore, the surface flatness of the sample and ablation stability can be well guaranteed to perform an accurate analysis. To evaluate the performance of the physical constraint, the LIBS spectra in four different detection forms including the physical constraint were acquired, and differences in their intensity, relative standard deviation (RSD) and signal-to-noise ratio (SNR) were compared. The comparison showed that good spectral quality can be obtained under a physical constraint. Finally, a prediction model based on partial least squares regression (PLSR) for ash and volatile matter in coal was established to verify the quantification performance. The results of model evaluation are very close to those of conventional modeling by ablating coal pellets, reflecting that the LIBS spectra under a physical constraint also enable good quantitative characterization for proximate analysis. Thus applying a physical constrain can effectively improve the online analysis capabilities of LIBS for coal quality in industrial fields.