Influence of variable selection on partial least squares discriminant analysis models for explosive residue classification

Using a series of thirteen organic materials that includes novel high-nitrogen energetic materials, conventional organic military explosives, and benign organic materials, we have demonstrated the importance of variable selection for maximizing residue discrimination with partial least squares discriminant analysis (PLS-DA). We built several PLS-DA models using different variable sets based on laser induced breakdown spectroscopy (LIBS) spectra of the organic residues on an aluminum substrate under an argon atmosphere. The model classification results for each sample are presented and the influence of the variables on these results is discussed. We found that using the whole spectra as the data input for the PLS-DA model gave the best results. However, variables due to the surrounding atmosphere and the substrate contribute to discrimination when the whole spectra are used, indicating this may not be the most robust model. Further iterative testing with additional validation data sets is necessary to determine the most robust model.     

Multivariate analysis of laser-induced breakdown spectroscopy chemical signatures for geomaterial classification

A large suite of natural carbonate, fluorite and silicate geological materials was studied using laser-induced breakdown spectroscopy (LIBS). Both single- and double-pulse LIBS spectra were acquired using close-contact benchtop and standoff (25 m) LIBS systems. Principal components analysis (PCA) and partial least squares discriminant analysis (PLS-DA) were used to identify the distinguishing characteristics of the geological samples and to classify the materials. Excellent discrimination was achieved with all sample types using PLS-DA and several techniques for improving sample classification were identified. The laboratory double-pulse LIBS system did not provide any advantage for sample classification over the single-pulse LIBS system, except in the case of the soil samples. The standoff LIBS system provided comparable results to the laboratory systems. This work also demonstrates how PCA can be used to identify spectral differences between similar sample types based on minor impurities.     

Laser-induced breakdown spectroscopy and chemometrics for classification of toys relying on toxic elements

Quality control of toys for avoiding children exposure to potentially toxic elements is of utmost relevance and it is a common requirement in national and/or international norms for health and safety reasons. Laser-induced breakdown spectroscopy (LIBS) was recently evaluated at authors' laboratory for direct analysis of plastic toys and one of the main difficulties for the determination of Cd, Cr and Pb was the variety of mixtures and types of polymers. As most norms rely on migration (lixiviation) protocols, chemometric classification models from LIBS spectra were tested for sampling toys that present potential risk of Cd, Cr and Pb contamination. The classification models were generated from the emission spectra of 51 polymeric toys and by using Partial Least Squares - Discriminant Analysis (PLS-DA), Soft Independent Modeling of Class Analogy (SIMCA) and K-Nearest Neighbor (KNN). The classification models and validations were carried out with 40 and 11 test samples, respectively. Best results were obtained when KNN was used, with corrected predictions varying from 95% for Cd to 100% for Cr and Pb.     

Detection of explosives in traces by laser induced breakdown spectroscopy: Differences from organic interferents and conditions for a correct classification

With the aim to study and to improve LIBS capability for detecting residues of energetic compounds in air surrounding, nine types of explosives and some potential interferents, placed in small quantities on a metallic support, were interrogated by a laser. Shot-to-shot behavior of the line intensities relative to the sample constituents was studied. The detected plasma was not stoichiometric and the line intensities, as well as their ratios, were changing even for an order of magnitude from one sampling point to another, particularly in the case of aromatic compounds. We explained some sources of such LIBS signal's behavior and this allowed us to establish a data processing procedure, which leads to a good linearization among the data sets. In this way, it was possible to determine some real differences between the LIBS spectra from explosives and interferents, and to correlate them with molecular formulas, with some known pathways for the molecule's decomposition and with successive chemical reactions in the plasma. Number spectral parameters, which distinguish the each studied explosive from other organic materials, were also determined and compared with previously published results relative to percentages of correct classifications for the same explosives. Experimental conditions for reliable recognition of the explosives by LIBS in air are also suggested, together with the parameters that should be considered or discarded from the classification procedure.     

Laser-induced breakdown spectroscopy for detection of explosives residues: a review of recent advances, challenges, and future prospects

In this review we discuss the application of laser-induced breakdown spectroscopy (LIBS) to the problem of detection of residues of explosives. Research in this area presented in open literature is reviewed. Both laboratory and field-tested standoff LIBS instruments have been used to detect explosive materials. Recent advances in instrumentation and data analysis techniques are discussed, including the use of double-pulse LIBS to reduce air entrainment in the analytical plasma and the application of advanced chemometric techniques such as partial least-squares discriminant analysis to discriminate between residues of explosives and non-explosives on various surfaces. A number of challenges associated with detection of explosives residues using LIBS have been identified, along with their possible solutions. Several groups have investigated methods for improving the sensitivity and selectivity of LIBS for detection of explosives, including the use of femtosecond-pulse lasers, supplemental enhancement of the laser-induced plasma emission, and complementary orthogonal techniques. Despite the associated challenges, researchers have demonstrated the tremendous potential of LIBS for real-time detection of explosives residues at standoff distances. Figure This review discusses the application of laser-induced breakdown spectroscopy (LIBS) to the problem of explosive residue detection. LIBS offers the capability for real-time, standoff detection of trace amounts of residue explosives on various surfaces     

Support vector machine classification of suspect powders using laser‐induced breakdown spectroscopy (LIBS) spectral data

Classification of suspect powders, by using laser‐induced breakdown spectroscopy (LIBS) spectra, to determine if they could contain Bacillus anthracis spores is difficult because of the variability in their composition and the variability typically associated with LIBS analysis. A method that builds a support vector machine classification model for such spectra relying on the known elemental composition of the Bacillus spores was developed. A wavelet transformation was incorporated in this method to allow for possible thresholding or standardization, then a linear model technique using the known elemental structure of the spores was incorporated for dimension reduction, and a support vector machine approach was employed for the final classification of the substance. The method was applied to real data produced from an LIBS device. Several methods used to test the predictive performance of the classification model revealed promising results. Published 2012. This article is a US Government work and is in the public domain in the USA.     

Sample classification for improved performance of PLS models applied to the quality control of deep-frying oils of different botanic origins analyzed using ATR-FTIR spectroscopy

The selection of an appropriate calibration set is a critical step in multivariate method development. In this work, the effect of using different calibration sets, based on a previous classification of unknown samples, on the partial least squares (PLS) regression model performance has been discussed. As an example, attenuated total reflection (ATR) mid-infrared spectra of deep-fried vegetable oil samples from three botanical origins (olive, sunflower, and corn oil), with increasing polymerized triacylglyceride (PTG) content induced by a deep-frying process were employed. The use of a one-class-classifier partial least squares-discriminant analysis (PLS-DA) and a rooted binary directed acyclic graph tree provided accurate oil classification. Oil samples fried without foodstuff could be classified correctly, independent of their PTG content. However, class separation of oil samples fried with foodstuff, was less evident. The combined use of double-cross model validation with permutation testing was used to validate the obtained PLS-DA classification models, confirming the results. To discuss the usefulness of the selection of an appropriate PLS calibration set, the PTG content was determined by calculating a PLS model based on the previously selected classes. In comparison to a PLS model calculated using a pooled calibration set containing samples from all classes, the root mean square error of prediction could be improved significantly using PLS models based on the selected calibration sets using PLS-DA, ranging between 1.06 and 2.91% (w/w).     

Advanced recognition of explosives in traces on polymer surfaces using LIBS and supervised learning classifiers

The large similarity existing in the spectral emissions collected from organic compounds by laser-induced breakdown spectroscopy (LIBS) is a limiting factor for the use of this technology in the real world. Specifically, among the most ambitious challenges of today's LIBS involves the recognition of an organic residue when neglected on the surface of an object of identical nature. Under these circumstances, the development of an efficient algorithm to disclose the minute differences within this highly complex spectral information is crucial for a realistic application of LIBS in countering explosive threats. An approach cemented on scatter plots of characteristic emission features has been developed to identify organic explosives when located on polymeric surfaces (teflon, nylon and polyethylene). By using selected spectral variables, the approach allows to design a concise classifier for alerting when one of four explosives (DNT, TNT, RDX and PETN) is present on the surface of the polymer. Ordinary products (butter, fuel oil, hand cream, olive oil and motor oil) cause no confusion in the decisions taken by the classifier. With rates of false negatives and false positives below 5%, results demonstrate that the classification algorithm enables to label residues according to their harmful nature in the most demanding scenario for a LIBS sensor.     

Multi-element and mineralogical analysis of mineral ores using laser induced breakdown spectroscopy and chemometric analysis

In the mining industry the quality and extent of an ore body is determined on the basis of routine assays conducted on drill core and chip samples. Both the elemental composition and the mineralogical classification are important in the characterisation of an ore body for commercial exploitation. Mining industry laboratories typically analyse large numbers of samples from both exploration and mine production environments.At CSIRO we have explored the application of chemometric methods of analysis in combination with laser induced breakdown spectroscopy (LIBS) in order to produce routine quantitative analysis of several ore types including iron, nickel and lead/zinc ores. In particular, principal components regression (PCR) has been applied to perform multi-element analysis of iron ore samples from Australia and West Africa. Calibration models for iron (4.8% Av. Relative Error), aluminium (2.2%), silicon (3.7%) and potassium (1.4%) were determined for the Australian ores. In addition phosphorous measurements were made at trace level for samples from West Africa (5.5% Av. Relative Error). LIBS measurements of segments of a nickel drill core were also analysed using PCR.Mineralogical classification using a combination of LIBS and principal components analysis (PCA) has also been explored. Broad discrimination of ore mineralogy was demonstrated on the basis of the PCA of LIBS spectra in selected emission wavelength bands. The combination of PCA and PCR offers potential for both broad mineralogical and elemental analysis for the minerals industry in exploration and in mine production for the on-line monitoring of ore quality.     

Classification of Greek Olive Oils from Different Regions by Machine Learning-Aided Laser-Induced Breakdown Spectroscopy and Absorption Spectroscopy

In the present work, the emission and the absorption spectra of numerous Greek olive oil samples and mixtures of them, obtained by two spectroscopic techniques, namely Laser-Induced Breakdown Spectroscopy (LIBS) and Absorption Spectroscopy, and aided by machine learning algorithms, were employed for the discrimination/classification of olive oils regarding their geographical origin. Both emission and absorption spectra were initially preprocessed by means of Principal Component Analysis (PCA) and were subsequently used for the construction of predictive models, employing Linear Discriminant Analysis (LDA) and Support Vector Machines (SVM). All data analysis methodologies were validated by both “k-fold” cross-validation and external validation methods. In all cases, very high classification accuracies were found, up to 100%. The present results demonstrate the advantages of machine learning implementation for improving the capabilities of these spectroscopic techniques as tools for efficient olive oil quality monitoring and control.     

Preliminary evaluation of laser induced breakdown spectroscopy for slurry samples

Laser-induced breakdown spectroscopy (LIBS) was applied to the analysis of simulant slurry samples used in the vitrification process of liquid radioactive wastes. A spectroscopic analysis was performed by two different detection systems: Czerny-Turner spectrometer coupled with intensified diode array detector (IDAD) and an Echelle spectrometer with intensified charge coupled device (ICCD). For the Czerny-Turner detection system, the normalized intensity method, which is the normalization of the atomic emission intensity by the released whole plasma emission area intensity, was employed to improve the reproducibility of LIBS signals. The Echelle detection system showed a high efficiency in simultaneous multi-element detection and determination of the physical quantities of the simulant.     

Parametric models to compute tryptophan fluorescence wavelengths from classical protein simulations

Fluorescence spectroscopy is an important method to study protein conformational dynamics and solvation structures. Tryptophan (Trp) residues are the most important and practical intrinsic probes for protein fluorescence due to the variability of their fluorescence wavelengths: Trp residues emit in wavelengths ranging from 308 to 360 nm depending on the local molecular environment. Fluorescence involves electronic transitions, thus its computational modeling is a challenging task. We show that it is possible to predict the wavelength of emission of a Trp residue from classical molecular dynamics simulations by computing the solvent‐accessible surface area or the electrostatic interaction between the indole group and the rest of the system. Linear parametric models are obtained to predict the maximum emission wavelengths with standard errors of the order 5 nm. In a set of 19 proteins with emission wavelengths ranging from 308 to 352 nm, the best model predicts the maximum wavelength of emission with a standard error of 4.89 nm and a quadratic Pearson correlation coefficient of 0.81. These models can be used for the interpretation of fluorescence spectra of proteins with multiple Trp residues, or for which local Trp environmental variability exists and can be probed by classical molecular dynamics simulations. © 2018 Wiley Periodicals, Inc.     

Evaluation of the effects of Candidatus Liberibacter asiaticus on inoculated citrus plants using laser-induced breakdown spectroscopy (LIBS) and chemometrics tools

This study investigated the organic and inorganic constituents of healthy leaves and Candidatus Liberibacter asiaticus (CLas)-inoculated leaves of citrus plants. The bacteria CLas are one of the causal agents of citrus greening (or Huanglongbing) and its effect on citrus leaves was investigated using laser-induced breakdown spectroscopy (LIBS) combined with chemometrics. The information obtained from the LIBS spectra profiles with chemometrics analysis was promising for the construction of predictive models to identify healthy and infected plants. The major, macro- and microconstituents were relevant for differentiation of the sample conditions. The models were then applied to different inoculation times (from 1 to 8 months). The models were effective in the classification of 82-97% of the diseased samples with a 95% significance level. The novelty of this method was in the fingerprinting of healthy and diseased plants based on their organic and inorganic contents.     

Elemental misinterpretation in automated analysis of LIBS spectra

In this work, the Stark effect is shown to be mainly responsible for wrong elemental allocation by automated laser-induced breakdown spectroscopy (LIBS) software solutions. Due to broadening and shift of an elemental emission line affected by the Stark effect, its measured spectral position might interfere with the line position of several other elements. The micro-plasma is generated by focusing a frequency-doubled 200 mJ pulsed Nd/YAG laser on an aluminum target and furthermore on a brass sample in air at atmospheric pressure. After laser pulse excitation, we have measured the temporal evolution of the Al(II) ion line at 281.6 nm (4s ¹ S-3p ¹ P) during the decay of the laser-induced plasma. Depending on laser pulse power, the center of the measured line is red-shifted by 130 pm (490 GHz) with respect to the exact line position. In this case, the well-known spectral line positions of two moderate and strong lines of other elements coincide with the actual shifted position of the Al(II) line. Consequently, a time-resolving software analysis can lead to an elemental misinterpretation. To avoid a wrong interpretation of LIBS spectra in automated analysis software for a given LIBS system, we recommend using larger gate delays incorporating Stark broadening parameters and using a range of tolerance, which is non-symmetric around the measured line center. These suggestions may help to improve time-resolving LIBS software promising a smaller probability of wrong elemental identification and making LIBS more attractive for industrial applications.     

基于多光谱特征融合技术的面粉掺杂定量分析方法

提出了一种基于拉曼光谱技术(Raman)和激光诱导击穿光谱技术(LIBS)的多光谱特征融合技术(MFFT),利用拉曼光谱中分子组分信息和激光诱导击穿光谱中原子组分信息之间的互补特性,采用自适应小波变换(AWT)-竞争性自适应加权(CARS)-偏最小二乘回归(PLS)建模技术,获取了面粉体系更为全面的特征信息。在多光谱特征融合技术中,首先采用AWT-CARS方法分别提取拉曼光谱和激光诱导击穿光谱中的特征变量,然后将两者的特征变量融合为一个向量,采用PLS方法构建MFFT模型,实现了面粉掺杂物的定量分析。通过对二氧化钛、硫酸铝钾等面粉掺杂体系建模分析,考察MFFT模型的有效性。结果表明,与单一拉曼光谱技术或激光诱导击穿光谱技术建立的预测模型相比,MFFT模型显著提升了模型的预测性能,二氧化钛和硫酸铝钾预测模型的线性相关系数分别从相对较差的Raman模型的0.884、0.877提升到0.981、0.980,其预测均方根误差分别从相对较差的Raman模型的0.151、0.154降低到0.069、0.068。表明多光谱特征融合技术可以准确提取Raman光谱中的分子信息和LIBS光谱中的元素信息,使其互为补充、互为校正,进而有效克服面粉基质对掺杂组分定量分析的干扰,显著提高模型的预测精度。     

Synergistic effect of the simultaneous chemometric analysis of H NMR spectroscopic and stable isotope (SNIF-NMR,O, C) data: Application to wine analysis

It is known that ¹H NMR spectroscopy represents a good tool for predicting the grape variety, the geographical origin, and the year of vintage of wine. In the present study we have shown that classification models can be improved when ¹H NMR profiles are fused with stable isotope (SNIF-NMR, ¹⁸O, ¹³C) data. Variable selection based on clustering of latent variables was performed on ¹H NMR data. Afterwards, the combined data of 718 wine samples from Germany were analyzed using linear discriminant analysis (LDA), partial least squares-discriminant analysis (PLS-DA), factorial discriminant analysis (FDA) and independent components analysis (ICA). Moreover, several specialized multiblock methods (common components and specific weights analysis (ComDim), consensus PCA and consensus PLS-DA) were applied to the data.     

Chemometrics‐enhanced laser‐induced thermal emission detection of PETN and other explosives on various substrates

Infrared emissions (IREs) of samples of pentaerythritol tetranitrate (PETN) deposited as contamination residues on various substrates were measured to generate models for the detection and discrimination of the important nitrate ester from the emissions of the substrates. Mid‐infrared emissions were generated by heating the samples remotely using laser‐induced thermal emission (LITE). Chemometrics multivariate analysis techniques such as principal component analysis (PCA), soft independent modeling by class analogy (SIMCA), partial least squares‐discriminant analysis (PLS‐DA), support vector machines (SVMs), and neural network (NN) were employed to generate the models for the classification and discrimination of PETN IREs from substrate thermal emissions. PCA exhibited less variability for the LITE spectra of PETN/substrates. SIMCA was able to predict only 44.7% of all samples, while SVM proved to be the most effective statistical analysis routine, with a discrimination performance of 95%. PLS‐DA and NN achieved prediction accuracies of 94% and 88%, respectively. High sensitivity and specificity values were achieved for five of the seven substrates investigated. Copyright © 2015 John Wiley & Sons, Ltd.     

Towards quantitative laser-induced breakdown spectroscopy analysis of soil samples

A quantitative analysis of chromium in soil samples is presented. Different emission lines related to chromium are studied in order to select the best one for quantitative features. Important matrix effects are demonstrated from one soil to the other, preventing any prediction of concentration in different soils on the basis of a univariate calibration curve. Finally, a classification of the LIBS data based on a series of Principal Component Analyses (PCA) is applied to a reduced dataset of selected spectral lines related to the major chemical elements in the soils. LIBS data of heterogeneous soils appear to be widely dispersed, which leads to a reconsideration of the sampling step in the analysis process.     

Multivariate calibration of spectra obtained by Laser Induced Breakdown Spectroscopy of plutonium oxide surrogate residues

Laser Induced Breakdown Spectroscopy (LIBS) was used to determine elemental concentration of plutonium oxide surrogate (cerium oxide) residue for monitoring the fabrication of lanthanide borosilicate glass. Quantitative analysis by LIBS is affected by the severe limitation of variation in the induced plasma due to changes in the matrix. Multivariate calibration was applied to LIBS data to predict the concentrations of Ce, Cr, Fe, Mo, and Ni. A total of 18 different samples were prepared to compare calibration from univariate data analysis and from multivariate data analysis. Multivariate calibration was obtained using Principal Component Regression (PCR) and Partial Least Squares (PLS). Univariate calibration was obtained from background-corrected atomic emission lines. Calibration results show improvement in the coefficient of determination from 0.87 to 0.97 for Ce compared to univariate calibration. The root mean square error also reduced from 7.46 to 2.93%. A similar trend was obtained for Cr, Fe, Mo, and Ni also. These results clearly demonstrate the feasibility of using LIBS for online process monitoring in a hazardous waste management environment.     

Transition-metal-based Chemosensing Ensembles: ATP Sensing in Physiological Conditions

Sensing of biologically relevant anionic substrates in physiological conditions, employing the strategy of the chemosensing ensembles, is reported. Coordination of a fluorescent indicator to a dicopper(II) polyazamacrocyclic receptor ([Cu²(L)]) results in the collapse of its fluorescence emission. Competitive binding of substrates for the receptor releases the indicator in solution, with full emission recovery. The spectral changes obtained for some indicators and substrates were analysed to determine their respective association constants for the receptor. Discrimination of micromolar ATP quantities from other interferents (small inorganic anions and well-known neurotransmitters) is improved by a judicious choice of the indicator, the resulting ATP sensor promising interesting biological applications.