Research On Non-metallic Elements Detection In Coal Using Laser-induced Breakdown Spectroscopy

As the main fuel of thermal power generation,coal plays an important role in the energy structure and the national economy in China.The non-metallic elements in coal have a crucial influence on the quality of coal.Therefore,it is of great significance for the rapid and accurate detection of non-metallic elements in coal.The traditional method for coal quality detection is difficult to meet the needs of modern industrial production due to the disadvantages of complex sample preparation,long time consuming and low accuracy.As an atomic emission spectroscopy technology,Laser-induced Breakdown Spectroscopy(LIBS)has become a hotspot in element analysis with the advantages of simple sample preparation,short detection time,and simultaneous detection of multiple elements.However,the low accuracy and sensitivity of the non-metallic elements in coal due to the weak spectral signal,overlapping spectral lines interference and background noise interference make LIBS unable to meet the needs of practical application for coal detection.In this regard,some quantitative analysis methods of carbon,hydrogen,nitrogen and sulfur in coal were studied in this thesis.The main research contents are as follows:The single-pulse LIBS system(SP-LIBS)and the double-pulse LIBS system(DP-LIBS)for the detection of non-metallic elements in coal were presented,respectively.By exploring the effects of sample preparation method,laser energy,defocusing distance,spectrometer acquisition delay and double pulse delay time on the spectral intensity and signal-to-noise ratio(SNR),the experimental parameters of SP-LIBS and DP-LIBS were optimized.To overcome the problem of low accuracy for the quantitative analysis of carbon,hydrogen,nitrogen and sulfur in coal,a multiple regression model based on partial least square regression(PLSR)was proposed to detect C,H,N and S elements in coal.For C,H,N and S elements,the predicted root-mean-square error(RMSEP)reached 0.5253,0.1023,0.0367,0.2067 wt.%,the average relative error(ARE)reached 0.747,2.2525,2.97,25.7wt.%,and R~2 reached 0.9967,0.9796,0.9667,0.8971,respectively.To overcome the problem of low accuracy and sensitivity for sulfur element detection in coal,the method of orthogonal DP-LIBS with re-heating mode in helium atmosphere was proposed,which enhanced the intensity and signal-to-noise ratio for S spectral lines and weakened the spectral lines interference.The basic calibration method was used to establish the calibration model for S element.The results showed that the limit of detection(Lo D)of S element can be as low as 0.038 wt.%,the root-mean-square error of cross-validation(RMSECV)can be as low as 0.1434 wt.%,and R~2 can be as high as 0.992.In summary,the LIBS quantitative analysis method of C,H,N and S elements in coal has been studied systematically.The PLSR model was introduced to reduce the matrix effect effectively;the analysis method of helium atmosphere combined with DP-LIBS was introduced to greatly improve the spectral intensity and signal-to-noise ratio,effectively reduce the interferences spectral lines.The quantitative analyses with high precision,high accuracy and high sensitivity for C,H,N and S elements in coal were achieved.