Benefiting from information-rich multi-frequency AC voltammetry coupled with chemometrics on the example of on-line monitoring of leveler component of electroplating bath

Simultaneous application of multiple sinusoidal waveforms perturbations superimposed onto DC staircase step significantly enriches current response. The measured current is characterized by a matrix of data rather than a conventional voltammetric output in a form of a vector. This increase of the dimensionality of the current response and therefore the wealth of analytical information is achieved without compromising the time of analysis. The natural approach for compression of such data and extraction of relevant information is by utilizing multi-way chemometric decomposition techniques. An electroplating solution presents a very challenging analyte for electroanalysis as its constituents interact synergistically with each other during both the plating process and its simulation during electroanalysis. For some components the mechanism is not entirely understood. Therefore, the only way to benefit from the analytical data is by employing soft modeling. The electrode processes involving additives rely heavily on adsorption and, indirectly, on electron transfer kinetics for which AC voltammetry is an analytical technique capable of delivering informative signals. This paper presents a rigorous universal method for calculating and validating an exemplary multi-way calibration of a leveler component in a copper electroplating bath used in the semiconductor industry. The method presented employs comparatively Parallel Factor Analysis coupled with Inverse Least Squares Regression and multi-linear Partial Least Squares. The calibration training set consists of multi-frequency AC voltammetric data subjected to pretreatments aiming to select informative independent variables and exclude outliers.     

Qualitative identification and quantitative comparison of Physochlainae Radix from different regions based on chemometric methods

Physochlainae Radix (PR) is an essential herbal medicine that has been generally applied for treating cough and asthma. In this study, a comprehensive strategy for quality evaluation of PR from different origins was established by integrating qualitative identification, quantitative analysis, and chemometric methods. A total of 58 chemical components were identified by ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight tandem mass spectrometry (UHPLC-Q-TOF-MS/MS), and a sensitive and rapid UHPLC-QqQ-MS/MS method was established for the simultaneous determination of 12 compounds. In addition, multivariate statistical analysis was applied for discriminant analysis to compare the differences among 30 batches of PR samples. The results showed that the 30 batches of PR collected from four provinces could be clustered into three categories, in which scoparone, protocatechuic acid, tropic acid, and scopolin were important components to distinguish the primary and non-primary producing areas, as well as superior and inferior products of PR. Chemometric results were consistent and validated each other, and systematically explained the intrinsic quality characteristics of PR. This study first demonstrated that LC-MS combined with multivariate statistical analysis, provided a comprehensive and effective means for quality evaluation of PR.     

低场核磁共振结合化学计量学方法快速检测掺假核桃油

以掺假核桃油样品为低场核磁共振检测对象,利用主成分分析法(PCA)和偏最小二乘回归法(PLSR)分析处理Carr-Purcell-Meiboom-Gill(CPMG)序列的核磁共振弛豫数据,旨在探求一种能快速检测核桃油品质的新方法。对几种常见掺假形式(掺入大豆油、玉米油、葵花油)的核桃油样品和纯核桃油样品进行检测和评价。实验结果表明:纯核桃油和掺入不同种类食用油的掺假核桃油在主成分得分图上可以得到很好的区分,且掺假样品随掺假比例在图中呈规律性分布;采用PLSR法对CPMG数据和实际掺假率进行回归,可实现对核桃油掺假水平的准确定量测定。方法快速、无损、准确,在食用油制品的品质控制及评价方面具有很大的应用潜力。     

基于内源性致香物质和化学计量学的烟草感官评价研究

采用主成分分析法结合遗传算法和神经网络,建立了基于烟草内源性致香物质的感官质量评价预测模型。利用气相色谱-质谱(GC-MS)技术对超临界萃取-分子蒸馏所得烟草精油中的内源性致香组分进行定性定量分析,汇总各类致香指标后,对其进行主成分分析;以提取所得5个主成分的得分作为输入变量,感官评吸分数作为输出变量,分别使用标准BP神经网络和遗传算法(GA)优化的BP神经网络建立预测模型。对比实验结果表明,GA优化后的模型预测效果更优,其预测值与实验值间的相关系数为0.96,预测均方根误差为1.81,说明GA-BP模型具有更好的拟合能力和预测能力,该模型能有效地预测烟草精油的感官品质。     

Quality assessment of crude and processed Arecae semen based on colorimeter and HPLC combined with chemometrics methods

Raw Arecae Semen, the seed of Areca catechu L., as well as Arecae Semen Tostum and Arecae semen carbonisata are traditionally processed by stir‐baking for subsequent use in a variety of clinical applications. These three Arecae semen types, important Chinese herbal drugs, have been used in China and other Asian countries for thousands of years. In this study, the sensory technologies of a colorimeter and sensitive validated high‐performance liquid chromatography with diode array detection were employed to discriminate raw Arecae semen and its processed drugs. The color parameters of the samples were determined by a colorimeter instrument CR‐410. Moreover, the fingerprints of the four alkaloids of arecaidine, guvacine, arecoline and guvacoline were surveyed by high‐performance liquid chromatography. Subsequently, Student's t test, the analysis of variance, fingerprint similarity analysis, hierarchical cluster analysis, principal component analysis, factor analysis and Pearson's correlation test were performed for final data analysis. The results obtained demonstrated a significant color change characteristic for components in raw Arecae semen and its processed drugs. Crude and processed Arecae semen could be determined based on colorimetry and high‐performance liquid chromatography with a diode array detector coupled with chemometrics methods for a comprehensive quality evaluation.     

A Metabolomic Approach to Beer Characterization

The consumers’ interest towards beer consumption has been on the rise during the past decade: new approaches and ingredients get tested, expanding the traditional recipe for brewing beer. As a consequence, the field of “beeromics” has also been constantly growing, as well as the demand for quick and exhaustive analytical methods. In this study, we propose a combination of nuclear magnetic resonance (NMR) spectroscopy and chemometrics to characterize beer. ¹H-NMR spectra were collected and then analyzed using chemometric tools. An interval-based approach was applied to extract chemical features from the spectra to build a dataset of resolved relative concentrations. One aim of this work was to compare the results obtained using the full spectrum and the resolved approach: with a reasonable amount of time needed to obtain the resolved dataset, we show that the resolved information is comparable with the full spectrum information, but interpretability is greatly improved.     

Development of a Wine Metabolomics Approach for the Authenticity Assessment of Selected Greek Red Wines

Wine metabolomics constitutes a powerful discipline towards wine authenticity assessment through the simultaneous exploration of multiple classes of compounds in the wine matrix. Over the last decades, wines from autochthonous Greek grape varieties have become increasingly popular among wine connoisseurs, attracting great interest for their authentication and chemical characterization. In this work, 46 red wine samples from Agiorgitiko and Xinomavro grape varieties were collected from wineries in two important winemaking regions of Greece during two consecutive vintages and analyzed using ultra-high performance liquid chromatography-quadrupole time-of-flight mass spectrometry (UHPLC-QToF-MS). A targeted metabolomics methodology was developed, including the determination and quantification of 28 phenolic compounds from different classes (hydroxycinnamic acids, hydroxybenzoic acids, stilbenes and flavonoids). Moreover, 86 compounds were detected and tentatively identified via a robust suspect screening workflow using an in-house database of 420 wine related compounds. Supervised chemometric techniques were employed to build an accurate and robust model to discriminate between two varieties.     

Inorganic Elements in Mytilus galloprovincialis Shells: Geographic Traceability by Multivariate Analysis of ICP-MS Data

The international seafood trade is based on food safety, quality, sustainability, and traceability. Mussels are bio-accumulative sessile organisms that need regular control to guarantee their safe consumption. However, no well-established and validated methods exist to trace mussel origin, even if several attempts have been made over the years. Recently, an inorganic multi-elemental fingerprint coupled to multivariate statistics has increasingly been applied in food quality control. The mussel shell can be an excellent reservoir of foreign inorganic chemical species, allowing recording long-term environmental changes. The present work investigates the multi-elemental composition of mussel shells, including Al, Cu, Cr, Zn, Mn, Cd, Co, U, Ba, Ni, Pb, Mg, Sr, and Ca, determined by inductively-coupled plasma mass-spectrometry in Mytilus galloprovincialis collected along the Central Adriatic Coast (Marche Region, Italy) at 25 different sampling sites (18 farms and 7 natural banks) located in seven areas. The experimental data, coupled with chemometric approaches (principal components analysis and linear discriminant analysis), were used to create a statistical model able to discriminate samples as a function of their production site. The LDA model is suitable for achieving a correct assignment of >90% of individuals sampled to their respective harvesting locations and for being applied to counteract fraud.     

基于角度度量法的混纺纤维定量检测

采用“基于角度度量的多变量回归方法”对维纶和腈纶混纺纤维各组分含量进行检测,并与直接用偏最小二乘法（PLS）对混和纤维的预测结果作对比。实验结果显示,PLS法对维纶和腈纶预测值与实际值的线性相关系数r均为0．9457,标准偏差为6．0906,均方根误差为6．9948。角度度量法对维纶预测值与实际值的线性相关系数r为0．9990,标准偏差为0．8929,均方根误差为2．1896;对腈纶预测值与实际值的线性相关系数r为0．9928,标准偏差为2．1896,均方根误差为3．9493。实验证明,角度度量法比PLS法更能准确表达定量关系,角度度量法可以显著降低分析操作对环境的要求,满足了近红外光谱在混纺纤维定量分析上的要求。     

Identification of markers of thermal processing (“roasting”) in aqueous extracts of Coffea arabica L. seeds through NMR fingerprinting and chemometrics

Roasting of Coffea arabica L. seeds gives rise to chemical reactions that produce more than 800 compounds, some being responsible for the desired organoleptic properties for which the beverage called “coffee” is known. In the industry, the “roasting profile,” that is, the times and temperatures applied, is key to influence the composition of roasted coffee beans and the flavour of the beverage made from them. The impact of roasting on the chemical composition of coffee has been the subject of numerous studies, including by nuclear magnetic resonance (NMR) spectroscopy. However, the roasting equipment and profiles applied in these studies are often far from real industrial conditions. In this work, the effects of two critical technological parameters of the roasting process, namely, the “development time” (the period of time after the “first crack,” a characteristic noise due to seed disruption) and the final roasting temperature on coffee extracts, were investigated. Seeds were roasted at pilot scale according to 13 industrial roasting profiles and extracted in D₂O. The extracts were analysed by ¹H NMR experiments. The NMR spectra were compared using (a) quantitative analysis of main signals by successive orders of magnitude and (b) chemometric tools (principal component analysis, partial least squares and sparse-orthogonal partial least squares analysis). This allowed to identify compounds, which may serve as markers of roasting and showed that changes in chemical composition can be detected even for slight change in final temperature (~1°C) or in total roasting time (~25 s).