Abstract: This paper deals with the interference of C and Fe plasma lines existing in the measurement of unburned carbon in fly ash by means of laser-induced breakdown spectroscopy ( LIBS) ,and explores the interference suppression based on the parameter optimization by experiments.In the investigation,first,according to the excitation and time evolution properties of C and Fe lines,the spectral line interference at various pulse energy and time delay are analyzed in a comparison way,and the optimized parameters are obtained.Then,two multiple regression models describing the unburned carbon are constructed respectively by using the peak intensity and the integrated intensity of elemental lines.Analytical results show that the interference between C and Fe lines at about 248nm can effectively be reduced by optimizing the pulse energy and the time delay; and that,in the optimized conditions,the intensity of Fe 247. 98nm line decreases,while C 247. 86nm line with high signal-to-noise ratio is obtained.Moreover,it is found that the multiple regression model constructed by peak intensity is reliable,with a regression coefficient r2 of 0. 999 and an intercept of 0. 005.

Key words: laser-induced breakdown spectroscopy, plasma, spectral line interference, laser energy, delay time, parameter optimization