Chemometrics and Numerical Methods in Libs

ISBN-13: 9781119759584 / Angielski / Twarda / 2022 / 368 str.

Vincenzo Palleschi

Spis treści

List of Contributors xiiiPreface xviiIntroduction and Brief Summary of the LIBS Development 1Part I Introduction to LIBS 51 LIBS Fundamentals 7Mohamad Sabsabi1.1 Interaction of Laser Beam with Matter 81.2 Basics of Laser-Matter Interaction 91.3 Processes in Laser-Produced Plasma 101.4 Factors Affecting Laser Ablation and Laser-Induced Plasma Formation 111.4.1 Influence of Laser Parameters on the Laser-Induced Plasmas 111.4.2 Laser Wavelength (lambda) 121.4.3 Laser Pulse Duration (tau) 121.4.4 Laser Energy (E) 131.4.5 Influence of Ambient Gas 131.5 Plasma Properties and Plasma Emission Spectra 14References 152 LIBS Instrumentations 19Mohamad Sabsabi and Vincenzo Palleschi2.1 Basics of LIBS instrumentations 192.2 Lasers in LIBS Systems 202.3 Desirable Requirements for Atomic Emission Spectrometers/Detectors 222.4 Spectrometers 232.4.1 Czerny-Turner Optical Configuration 232.4.2 Paschen-Runge Design 242.4.3 Echelle Spectrometer Configuration 252.5 Detectors 262.5.1 Photomultiplier Detectors 262.5.2 Solid-State Detectors 272.5.3 The Interline CCD Detectors 272.5.3.1 The Image Intensifier 28References 293 Applications of LIBS 31Vincenzo Palleschi and Mohamad Sabsabi3.1 Industrial Applications 313.1.1 Metal Industry 313.1.2 Energy Production 343.2 Biomedical Applications 343.3 Geological and Environmental Applications 363.4 Cultural Heritage and Archaeology Applications 373.5 Other Applications 37References 38Part II Simplications of LIBS Information 454 LIBS Spectral Treatment 47Sabrina Messaoud Aberkane, Noureddine Melikechi and Kenza Yahiaoui4.1 Introduction 474.2 Baseline Correction 474.2.1 Polynomial Algorithm 484.2.2 Model-free Algorithm 494.2.3 Wavelet Transform Model 524.3 Noise Filtering 554.3.1 Wavelet Threshold De-noising (WTD) 554.3.2 Baseline Correction and Noise Filtering 594.4 Overlapping Peak Resolution 604.4.1 Curve Fitting Method 614.4.2 The Wavelet Transform 644.5 Features Selection 664.5.1 Principal Component Analysis 684.5.2 Genetic Algorithm (GA) 684.5.3 Wavelet Transformation (WT) 68References 715 Principal Component Analysis 81Mohamed Abdel-Harith and Zienab Abdel-Salam5.1 Introduction 815.1.1 Laser-Induced Breakdown Spectroscopy (LIBS) 815.2 The Principal Component Analysis (PCA) 825.3 PCA in Some LIBS Applications 835.3.1 Geochemical Applications 835.3.2 Food and Feed Applications 855.3.3 Microbiological Applications 885.3.4 Forensic Applications 915.4 Conclusion 94References 946 Time-Dependent Spectral Analysis 97Fausto Bredice, Ivan Urbina, and Vincenzo Palleschi6.1 Introduction 976.2 Time-Dependent LIBS Spectral Analysis 986.2.1 Independent Component Analysis 986.2.2 3D Boltzmann Plot 1026.2.2.1 Principles of the Method 1036.3 Applications 1096.3.1 3D Boltzmann Plot Coupled with Independent Component Analysis 1096.3.2 Analysis of a Carbon Plasma by 3D Boltzmann Plot Method 1096.3.3 Assessment of the LTE Condition Through the 3D Boltzmann Plot Method 1146.3.4 Evaluation of Self-Absorption 1146.3.5 Determination of Transition Probabilities 1186.3.6 3D Boltzmann Plot and Calibration-free Laser-induced Breakdown Spectroscopy 1216.4 Conclusion 123References 123Part III Classification by LIBS 1277 Distance-based Method 129Hua Li and Tianlong Zhang7.1 Cluster Analysis 1327.1.1 Introduction 1327.1.2 Theory 1337.1.2.1 K-means Clustering 1337.1.2.2 Hierarchical Clustering 1347.1.3 Application 1357.2 Independent Components Analysis 1387.2.1 Introduction 1387.2.2 Theory 1387.2.3 Application 1407.3 K-Nearest Neighbor 1437.3.1 Introduction 1437.3.2 Theory 1437.3.3 Application 1457.4 Linear Discriminant Analysis 1457.4.1 Introduction 1457.4.2 Theory 1487.4.2.1 The Calculation Process of LDA (Two Categories) 1487.4.3 Application 1517.5 Partial Least Squares Discriminant Analysis 1537.5.1 Introduction 1537.5.2 Theory 1557.5.3 Application 1577.6 Principal Component Analysis 1617.6.1 Introduction 1617.6.2 Theory 1647.6.3 Application 1667.7 Soft Independent Modeling of Class Analogy 1747.7.1 Introduction 1747.7.2 Theory 1757.7.3 Application 1777.8 Conclusion and Expectation 180References 1818 Blind Source Separation in LIBS 189Anna Tonazzini, Emanuele Salerno, and Stefano Pagnotta8.1 Introduction 1898.2 Data Model 1938.3 Analyzing LIBS Data via Blind Source Separation 1938.3.1 Second-order BSS 1938.3.2 Maximum Noise Fraction 1948.3.3 Independent Component Analysis 1968.3.4 ICA for Noisy Data 1978.4 Numerical Examples 1978.5 Final Remarks 206References 2079 Artificial Neural Networks for Classification 213Jakub Vrábel, Erik Képes, Pavel PoYízka, and Jozef Kaiser9.1 Introduction and Scope 2139.2 Artificial Neural Networks (ANNs) 2149.3 Cost Functions and Training 2169.4 Backpropagation 2199.5 Convolutional Neural Networks 2219.6 Evaluation and Tuning of ANNs 2249.7 Regularization 2279.8 State-of-the-art LIBS Classification Using ANNs 2299.9 Summary 233Acknowledgments 234References 23410 Data Fusion: LIBS + Raman 241Beatrice Campanella and Stefano Legnaioli10.1 Introduction 24110.2 Data Fusion Background 24210.3 Data Treatment 24410.4 Working with Images 24510.4.1 Vectors Concatenation 24610.4.2 Vectors Co-addition 24610.4.3 Vectors Outer Sum 24610.4.4 Vectors Outer Product 24710.4.5 Data Analysis 24710.5 Applications 24810.6 Conclusion 253References 253Part IV Quantitative Analysis 25711 Univariate Linear Methods 259Stefano Legnaioli, Asia Botto, Beatrice Campanella, Francesco Poggialini, Simona Raneri, and Vincenzo Palleschi11.1 Standards 25911.2 Matrix Effect 26011.3 Normalization 26111.4 Linear vs. Nonlinear Calibration Curves 26411.5 Figures of Merit of a Calibration Curve 26711.5.1 Coefficient of Determination 27011.5.2 Root Mean Squared Error of Calibration 27011.5.3 Limit of Detection 27011.6 Inverse Calibration 27311.7 Conclusion 274References 27412 Partial Least Squares 277Zongyu Hou, Weiran Song, and Zhe Wang12.1 Overview 27712.2 Partial Least Squares Regression Algorithms 27812.2.1 Nonlinear Iterative PLS 27812.2.2 SIMPLS Algorithm 27912.2.3 Kernel Partial Least Squares 27912.2.4 Locally Weighted Partial Least Squares 28012.2.5 Dominant Factor-based Partial Least Squares 28112.3 Partial Least Squares Discriminant Analysis 28212.4 Results of Partial Least Squares in LIBS 28312.4.1 Coal Analysis 28312.4.2 Metal Analysis 28512.4.3 Rocks, Soils, and Minerals Analysis 28512.4.4 Organics Analysis 29112.5 Conclusion 291References 29513 Nonlinear Methods 303Francesco Poggialini, Asia Botto, Beatrice Campanella, Stefano Legnaioli, Simona Raneri, and Vincenzo Palleschi13.1 Introduction 30313.2 Multivariate Nonlinear Algorithms 30413.2.1 Artificial Neural Networks 30413.2.1.1 Conventional Artificial Neural Networks 30413.2.1.2 Convolutional Neural Networks 31013.2.2 Other Nonlinear Multivariate Approaches 31213.2.2.1 The Franzini-Leoni Method 31213.2.2.2 The Kalman Filter Approach 31313.2.2.3 Calibration-Free Methods 31413.3 Conclusion 315References 31614 Laser Ablation-based Techniques - Data Fusion 321Jhanis Gonzalez14.1 Introduction 32114.2 Data Fusion of Multiple Analytical Techniques 32214.2.1 Low-level Fusion 32214.2.2 Mid-level Fusion 32314.2.3 High-level Fusion 32414.3 Data Fusion of Laser Ablation-Based Techniques 32414.3.1 Introduction 32414.3.2 Classification of Edible Salts 32614.3.2.1 LIBS and LA-ICP-MS Measurements of the Salt Samples 32714.3.2.2 Mid-Level Data Fusion of LIBS and LA-ICP-MS of Salt Samples 32714.3.2.3 PLS-DA Classification Model for Salt Samples 33314.3.3 Coal Discrimination Analysis 33414.3.3.1 LIBS and LA-ICP-TOF-MS Measurements of the Coal Samples 33514.3.3.2 Mid-Level Data Fusion of LIBS and LA-ICP-TOF-MS of Coal Samples 33514.3.3.3 PCA Combined with K-means Cluster Analysis for Coal Samples 33814.3.3.4 PLS-DA and SVM for Coal Samples Analysis 34014.4 Comments and Future Developments 341Acknowledgments 343References 343Part V Conclusions 34715 Conclusion 349Vincenzo PalleschiIndex 351

O autorze

Vincenzo Palleschi is a Senior Researcher with the Institute of Chemistry of Organometallic Compounds, Italian Research Council and Professor of Advanced Analytical Chemistry at the University of Pisa. He is Associate Editor of the Journal of Advanced Research and a member of the Editorial Advisory Boards of Spectrochimica Acta B and Reviews in Analytical Chemistry. He has published more than 140 scientific papers on LIBS, making him the most productive author in LIBS ever. His paper on Calibration-Free LIBS, published in 1999, is the most quoted research paper in LIBS. In 2000 he was the organizer and chairperson of the First International Conference on LIBS in Pisa (Italy).