Zhimin Lu

Correction of C–Fe line interference for the measurement of unburned carbon in fly ash by LIBS

For the serious interference between the Fe 247.98 nm and the C 247.86 nm lines in the analysis of fly ash when using a poor-resolution spectrograph, this paper presents a correction method that uses Fe lines to correct the C–Fe line interference. The integrated intensity of the spectral lines is chosen as the input intensity to obtain the intensity ratio of non-interference Fe lines and the Fe 247.98 nm line by LIBS measurements, and to obtain the C 247.86 nm line integrated intensity from the overlapping peak. Fe 248.33 nm, Fe 250.11 nm, and Fe 251.08 nm lines were used as correction lines to predict the unburned carbon (UC) content in fly ash. We applied multivariate calibration methods to establish the calibration curve of UC in the measured samples, and used quantitative analysis to further verify the effect of Fe lines as correction lines on the detection accuracy of UC in fly ash by LIBS. The results show that the correlation coefficients R2 of multivariate calibration all have improved based on Fe 248.33 nm, Fe 250.11 nm, and Fe 251.08 nm line correction. In terms of the overall evaluation from the calibration curve fitting and predictive accuracy of quantitative analysis, the Fe 248.33 nm line has the best performance as a correction line to correct C–Fe line interference when measuring UC content in fly ash. The correlation coefficients R2 of the multivariate calibration and the averaged prediction absolute error of the UC content are 0.998 and 0.002 wt%, respectively, which is significantly improved compared with that without correction.

Optimizing the binder percentage to reduce matrix effects for the LIBS analysis of carbon in coal

Quantitative analysis of elements by laser-induced breakdown spectroscopy (LIBS) is significantly affected by matrix effects in coal. Coal powder was mixed with the KBr binder and pressed into pellets to reduce matrix effects. Four groups of mixed-pressed pellet samples were prepared from nine different coal types and different percentages of the KBr binder (KBr accounts for 0 wt%, 30 wt%, 60 wt%, and 90 wt%, respectively). To optimize the percentages of the KBr binder in the mixed-pressed pellets, the influence of KBr binder on laser-induced plasma was investigated in detail. The results indicate that the plasma excitation temperature decreases with increase of the KBr binder concentration. The difference of the excitation temperature between the nine different coal samples was minimal when KBr accounts for 60 wt% in the mixed-pressed pellets. The relative standard deviation of the excitation temperature is 4.26%. The matrix changes from coal to KBr when the percentages of KBr are higher than 60 wt%, which was confirmed by scanning electron microscopy images of the ablated crater. Finally, Si and K were individually chosen as the internal calibration element to construct the calibration curves of carbon. Better results were obtained when Si I 288.16 nm was used as the internal standard. The correlation coefficients R2 of the four groups of mixed-pressed pellet samples are 0.835, 0.893, 0.983, and 0.903, respectively. Hence, the appropriate percentage of binders needs to be carefully confirmed to reduce matrix effects in quantitative analysis of coal by LIBS.

Optimizing critical parameters for the directly measurement of particle flow with PF-SIBS

A novel measurement technology named as particle flow-spark induced breakdown spectroscopy (PF-SIBS) was reported for real-time measurement of solid materials. Critical measurement parameters of PF-SIBS were optimized and a set of fly ashes with different carbon content were measured for evaluation of measurement performance. Four electrode materials, tungsten, copper, molybdenum and platinum, were compared in the aspects of signal stability, line interference and electrode durability. Less line interference and better signal stability were obtained with W and Cu electrode, while W electrode has better durability. Quartz sand with diameters from 48 μm to 180 μm were tested to investigate the influence of particle size. As the particle diameter increased, the intensity of Si 288.16 nm line decreased while that of ambient air constituents increased. To reduce the particle effect, the sum intensity from sample and ambient air were introduced to correct. The RSD of line intensity between the five diameters were reduced from 67.30% to 16.59% with Cu electrodes and from 63.21% to 13.64% with W electrodes. With the optimal measurement parameters and correction, fly ash samples with different carbon content were tested and the correlation coefficients R2 of multivariate calibration achieved 0.987.

Real-time measurement of constituents in solid materials using particle flow spark induced breakdown spectroscopy

This paper presents an analytical scheme that combines spark-induced breakdown spectroscopy with particle flow (PF-SIBS) for the real-time measurement of the constituents in powdered materials. With high voltage power supply instruments, a vibrational feeder and a spectrometer, the measuring system is low-cost and robust. To demonstrate its applicability, particle samples with varied carbon content were tested in LIBS and PF-SIBS for the comparison of their relative standard deviation (RSD), false spectra ratio, signal-to-noise ratio (SNR) and calibration results. The RSD of the C I 247.86 nm line in LIBS ranged from 35.72% to 62.51% for eight samples, while for PF-SIBS it ranged from 19.63% to 46.57%. The SNR of the Si I 288.15 nm was set as the index for the partial breakdown spectral identification and about 10% of the spectra were identified as false in LIBS, while all the spectra were identified as true spectra in PF-SIBS. The SNR of the C line in PF-SIBS was enhanced by 4 times as compared with LIBS. Since lines emitted from the tungsten electrode are inevitable, a correction method was employed to correct the interference between the C I 247.86 nm and W II 247.77 nm lines. The correlation coefficient, R2, of the linear multivariate calibration of LIBS was 0.957, while that of PF-SIBS was greater than 0.990 and can reach 0.995 after C–W interference correction. With interference correction in PF-SIBS, the average absolute error of the sample validation was 0.38% and the limit of detection (LOD) was 0.46%. These results demonstrate the feasibility of PF-SIBS as a powerful detection technology for the real-time measurement of the constituents in powdered materials and the interference correction can improve the measurement effect at low concentration.

Identifying laser-induced plasma emission spectra of particles in a gas–solid flow based on the standard deviation of intensity across an emission line

A new conditional data processing scheme named the standard deviation (SD) method is presented and evaluated for identifying the spectral data of a gas–solid flow based on laser-induced breakdown spectroscopy. The SD method is compared with two conditional data processing methods called the signal-to-noise ratio (SNR) method and the absolute peak intensity method. First, the performance of the three methods for identifying the spectral data of the same fly ash sample was compared. Then, the stability of the three methods was checked by identifying the spectral data of a set of 12 coal samples under different experimental conditions. The rejection rate, false rejection rate and false acceptance rate under various conditional analysis threshold values were used to evaluate these three different methods. The characteristic peaks at Si 288.16 nm and C 247.86 nm were selected for the analysis of fly ash and coal samples, respectively. The results show that true data and spurious data could be completely and accurately identified by the SD method. Moreover, it has been proved that the threshold values of the absolute peak intensity method and the SNR method fluctuate dramatically while the threshold value of the SD method remains stable under different experimental conditions. Compared with the other two methods, the SD method has better applicability and reliability when faced with variable detection conditions. So it has a greater advantage in identifying spurious laser-induced plasma emission spectra of particles in a gas–solid flow.

Development of a Rapid Coal Analyzer Using Laser-Induced Breakdown Spectroscopy (LIBS):

Determination of coal quality plays a major role in coal-fired power plants and coal producers for optimizing the utilization efficiency and controlling the quality. In this work, a rapid coal analyzer based on laser-induced breakdown spectroscopy (LIBS) was developed for rapid quality analysis of pulverized coal. The structure of the LIBS apparatus was introduced in detail. To avoid time-consuming and complicated sample preparation, a pulverized feeding machine was designed to form a continuously stable coal particle flow. The standard deviation (SD) of characteristic peaks was used to estimate the spectral valid data in this experiment. Coupled with cluster analysis, artificial neural networks and genetic algorithm are employed as a nonlinear regression method in order to indicate the relationship between coal quality and the corresponding plasma spectra. It is shown that the average absolute error of ash, volatile matter, fixed carbon, and gross calorific value for the validation set is 0.82%, 0.85%, 0.96%, and 0.48 MJ/kg. The average standard deviation of repeated samples is 1.64%, 0.92%, 1.08%, and 0.86 MJ/kg, showing a high sample-to-sample repeatability. This rapid coal analyzer is capable of performing reliable and accurate analysis of coal quality.