排序方式: 共有9条查询结果,搜索用时 18 毫秒
1
1.
应用中国地质大学(武汉)珠宝学院自制的激光诱导离解光谱(LIBS)设备对5块已知成分和32块未知成分的翡翠进行测量,采用优化外标法分析了LIBS测试稳定性与测量精度,测试计算结果平均值与真值偏差小于1%,测试计算结果标准偏差优于1.2%;通过测试结果计算待测样品的晶体化学通式,其不同晶位阳离子系数总数之比为(O.913±1.075):(0.902±1.182):(1.909±2.087),基本上符合翡翠阳离子晶位系数1;1;2的理论值,验证了LIBS定量区分硬玉与绿辉石的可行性。试验样品的LIBS定量分析结果显示,硬玉与绿辉石中SiO2的质量分数基本上稳定在59%左右,硬玉中Na2O和Al2O。的质量分数分别在13%和23%左右,CaO,FeO2r,MgO的总质量分数低于4%,而绿辉石中Na2O和Al2O。的质量分数比硬玉的低,翡翠中CaO,FeO2,MgO的总质量分数随着绿辉石的质量分数增加而增大,可达到27%。此外,绿辉石中Na20和CaO的质量分数呈现很好的负线性相关,相关系数r1=--0.9901;Al2O2与MgO的质量分数也呈现很好的负线性相关,相关系数r2=-0.9666。 相似文献
2.
3.
4.
Laser-induced breakdown spectroscopy with artificial neural network processing for material identification 总被引:1,自引:0,他引:1
Laser-induced breakdown spectroscopy (LIBS) has demonstrated its high potential in measurement of material composition in many areas including space exploration. LIBS instruments will be parts of payloads for the 2011 Mars Science Laboratory NASA-led mission and the ExoMars mission planned by ESA. This paper considers application of artificial neural networks (ANN) for material identification based on LIBS spectra that may be obtained with a portable instrument in ambient conditions. The several classes of materials used in this study included those selected to represent the sites analogues to Mars. In addition, metals and aluminum alloys were used to demonstrate ANN capabilities. Excellent material classification is achieved with single-shot measurements in real time. 相似文献
5.
岩石样品中复杂的基质效应严重影响激光诱导击穿光谱(LIBS)定量分析的准确性,其原因是目标元素的发射特性会受到基质的影响,导致其发射强度偏离理想的规律。为提高定量分析准确性,本文提出一种基于岩性基质特性的预分类定量分析方法。该方法首先构建基于k近邻(kNN)与支持向量机(SVM)算法的多层分类模型识别样品的岩性进行分类,通过kNN算法将样品分成碳酸盐和硅酸盐两大类,再利用SVM算法将大类细分成6类,而后针对不同岩性样品分别构建元素定量模型。通过采用预分类方法,可以确保分析的样品具有相似的化学成分,更好地确定分析时的基准线和校准曲线,从而减少分析中的不确定度,提高定量准确性。kNN算法通过交叉验证选取最优的k值,同时使用网格寻优方法确定了SVM算法中关键惩罚参数C和RBF宽度参数γ,利用该分类模型对来自6类岩性的39个国标岩石样品和国标岩石混合样品中的Si、Ca、Mg和K元素进行分析,岩性识别的准确率达100%,保证了后续定量分析的准确性,并针对不同岩性中的不同元素采用了合适的预处理方式提升光谱数据的稳定性。相比于传统标准曲线定量方法,采用预分类方法可以减少不同岩性基质之间的相互影响,从而减小样品基质非均匀性带来的误差。对比两种方法进行数据分析,测试集样品的预测值与参考值相关性分析系数从0.231~0.664提高至0.994~0.999,平均相对标准偏差从38.2%降低至8.6%。与传统定量分析方法相比较,采用预分类定量分析方法所构建模型对上述4种元素定量分析结果准确性有着明显的提高,为提高岩石元素定量分析准确性提供新的思路,拓宽了LIBS技术的实际应用范围。 相似文献
6.
7.
利用常规宝石学测试、X射线荧光光谱仪(XRF)、激光诱导离解光谱仪(LIBS)、电子探针(EPMA)对近来在珠宝市场上的铅玻璃充填碧玺样品进行测试与分析,旨在探讨其鉴定特征。结果表明,该铅玻璃充填样品的充填特征不明显,与传统的有机物充填处理方法不同;X射线荧光光谱和激光诱导离解光谱的测试结果显示,该铅玻璃充填碧玺样品具有明显的Pb峰。铅玻璃充填碧玺样品中充填物的电子探针二次电子像及背散射电子像特征,可作为碧玺样品是否经过了铅玻璃充填的诊断性依据,其充填物的化学成分主要为Si和Pb,呈不规则斑块状分布于裂隙中。 相似文献
8.
KArMars: A Breadboard Model for In Situ Absolute Geochronology Based on the K–Ar Method Using UV‐Laser‐Induced Breakdown Spectroscopy and Quadrupole Mass Spectrometry
下载免费PDF全文
![点击此处可从《Geostandards and Geoanalytical Research》网站下载免费的PDF全文](/ch/ext_images/free.gif)
Damien Devismes Pierre‐Yves Gillot Jean‐Claude Lefèvre Claire Boukari Francis Rocard Florence Chiavassa 《Geostandards and Geoanalytical Research》2016,40(4):517-532
We present a breadboard prototype to perform in situ dating applicable to planetary exploration. Based on the K–Ar dating method and using instruments inspired by flight‐proven analytical components, ‘KArMars’ ablated a geological sample under high vacuum with a quadrupled ultraviolet (UV at 266 nm) Nd:YAG laser. During ablation, the K content of the target material was given by laser‐induced breakdown spectroscopy and the released 40Ar was measured with a quadrupole mass spectrometer. Because K was measured as a concentration and 40Ar as a count of atoms, these values were converted using the ablated mass given by the product of the density and the ablated volume. The uncertainties of the age measurement were < 15%. The quality of the K–Ar measurements was enhanced by the advantages of UV laser ablation such as the minimisation of thermal effects on argon diffusion. This work demonstrates that a specialised instrument inspired by this set‐up could provide in situ absolute geochronology with sufficient precision for scientific investigations, particularly where the crater density counting provides higher uncertainties on Mars. 相似文献
9.
岩石岩性识别在油气田探测开发、研究地球成因及演化发展、地质灾害分析预测等众多方面起着不可替代的导向作用,因此岩石的识别分类对于地质勘探分析来说至关重要。为了提高岩石的分类准确率,提出了一种基于激光诱导击穿光谱技术(LIBS)的岩石表面指纹图谱分析及分类方法。通过LIBS对岩石表面不同位置进行激发,获取原始光谱数据。对收集到的光谱数据进行去除异常点、归一化等预处理操作,根据岩石矿物成分确定五种含量差异较大元素(硅、铝、钾、钠、镁)的特征谱线并得到元素指纹图谱。然后选择支持向量机(SVM)作为分类器进行分类,分别建立利用光谱均值的分类模型和多维指纹图谱融合的分类模型,并对两种分类结果进行比较。利用光谱均值的分类模型准确率为59.4%,多维指纹图谱融合的模型分类准确率为96.5%。实验结果表明,元素指纹图谱展示了岩石表面元素分布,可以充分利用不同种类岩石本身的不均匀性结构信息,极大地提高了岩石的分类准确率。 相似文献
1