| Packed microgranular matters are ubiquitous in many industrial activities,and their chemical composition is a proximate witness of the processes of corresponding material’s production,processing and treatment.Therefore,convenient and fast analysis for the packed microgranular matters will play a crucial role in the quality evaluation of products and the dynamic control of corresponding industrial processes.In addition,the Driven Advanced Nuclear Energy System(ADANES)originally proposed by Chinese Academy of Sciences is also facing the challenge of in-situ detection of randomly packed spent fuel.Therefore,in-situ detection of such packed microgranular matters is of great significance not only to reduce costs and increase efficiency in industrial production,but also to the sustainable development of nuclear energy!However,traditional techniques for analyzing the chemical composition of packed microgranular matters are dilatory,labor-intensive,destructive to samples,and unsuitable for in-situ and rapid analysis.Recent studies have confirmed that laser-induced breakdown spectroscopy(LIBS)technique has the potential to in-situ and rapidly analyze the packed microgranular matters.It should be mentioned that the“in-situ”and“rapidly”advantages of LIBS technique is based on the fact that any pretreatment for the sample to be tested is not required.Therefore,when directly analyzing packed microgranular matters,their grain size and packing fraction are the source of main physical matrix effects that affect the analytical performance of the technique.Therefore,clarifying how these two physical parameters affect the analytical performance and uncovering the underlying physics are key scientific issues that needs to be solved before industrial application of LIBS technique for packed microgranular matters.In this dissertation,based on the application requirements and the research background mentioned above,a systematic study was conducted on packed microgranular matters with different grain sizes and packing fractions using a diagnostic approach that combines spectroscopy,dynamic and static images.The evolution behaviors of LIBS signal with the two sample’s parameters were measured and their underlying physical properties were revealed.Furthermore,a simulation experiment was conducted to directly analyze microelements in packed microgranular samples using LIBS technique.The main research contents and conclusions obtained in this dissertation are summarized as follows:(1)In order to study the evolution behavior of LIBS signal with grain size,a systematic experiment was carried out with a focused nanosecond laser(1064 nm,7 ns)to ablate the surface of copper packed microgranular samples with different grain sizes(49～390μm).It is found that the size dependent LIBS shows a novel critical behavior(critical size approximately 100μm).Taking the packed microgranular samples as a type of soft matter with non-Newtonian fluid properties,The observed critical behavior can be explained as follows:in the range of the grain size above(below)the critical value,the packed microgranular sample has a yield stress larger(smaller)than the shock pressure imparted by the energetic processes of the laser-induced plasma generation and expansion,and thus,behaves like an elastic solid(a viscous fluid)to assist(impede)the formation of the plasma with high temperature and high density as an optical emission source for spectrochemical analysis.In addition,the measured trends of the LIBS signal versus the laser fluence(12.4～35.4 J/cm~2)for four cupper packed microgranular samples with grain sizes larger than the critical size suggest the identification of three laser fluence ranges,where the size effect plays a different role.This experimental phenomenon is also well explained by the packed microgranular samples as a type of soft matter.This study has practical significance in developing the LIBS technique for analyzing packed microgranular matters,i.e.,it is possible to greatly reduce or event eliminate the size effect on the analytical performance by limiting the grain size and laser fluence in suitable ranges.(2)In order to study the evolution behavior of LIBS signal with volume fraction,a systematic experiment was carried out with a focused nanosecond laser(1064 nm,7ns)to ablate the packed microgranular samples prepared by random packings of(72±20μm)copper grains into various volume fractions ranging from 0.593 to 0.623.It is found that,with the increase of volume fraction,the emission intensities of neutral copper lines increase and those of ionic copper lines remain almost unchanged.Taking the packed microgranular samples as a type of soft matter,the observations can be explained by introducing a detailed surface absorption mechanism during interaction between the laser-induce plasma and the soft surface after the end of a laser pulse.Specifically,during the expansion process of the plasma generated from the soft sample surface,the surface will absorb the kinetic energy and particles themselves(the worse the surface mechanical properties,the higher the absorption efficiency)near the surface end of plasma.This study provides some guidelines on how to choose specific analytical lines at which LIBS as an analytical technique to quantify elements embedded in soft materials is viable without considering the difference in surface mechanical properties.Finally,based on this detailed absorption mechanism introduced here,we proposed a new method of characterizing the surface mechanical property of soft materials via LIBS technique.(3)A specific experiment was carried out to demonstrate the feasibility of quantitative analysis of mixed packed microgranular matters using LIBS technique.Nine sets of binary mixed packed microgranular samples with doping concentrations ranging from 0.1%to 25 wt.%were prepared using copper grains(matrix)and Cr-Fe grains(dopant)with the same grain size(72±20μm)From the measurements on LIBS signal of these mixed packed microgranular samples we found that there is a laser pulse sampling rate similar to the case of aerosol LIBS,namely a doping-concentration dependent pulse sampling rate.We established firstly the relationship between the pulse sampling rate versus the doping concentration by introducing a parameter of effective ablation radius.Based on the established relationship,we developed the univariate and multivariate calibration models for the samples using various accumulating number of single-shot LIBS.It is found that,for such mixed packed microgranular samples,the prediction performance of the multivariate calibration model is much better than that of the multivariate one.It is attributed to that such a special pulse sampling rate behavior seriously influence the selection of the features having a strong correlation with the doping concentration under the establishment of multivariate calibration model. |
PDF Full Text Request