| The arrival of the "data tsunami", brings about the opportunities for theresearchers of chemometrics, and also produces a huge challenge. A lot of complexhuge instrument data has appeared, however, how to extract a small amount ofinterested information from the oceans of data is the main challenge of the research ersof chemometrics. A lot of studies are needed for the development of related methodsthat can handle the complex multi-way data. Fortunately, with years of development,the multi-way calibration method has been successful applied in analytical chemistry,environment, food hygiene, medical research and other fields. Based on the existenceof the “second order or higher order advantage”, it has become an interestingalternative to deal with the complex system for the quantitative analysis. Theexperimental data can be directly resolved to the spectra of each contributedcomponent by means of “mathematical separation”. We believe that in the near future,as an important component of chemometrics, multi-way calibration methods will play amuch stronger power to solve the complex problems of life science closely related tohuman. From reality, the authors have studied the multi-way calibration theory and itsrelated applications in this article, mainly including the following two aspects:Firstly, the first part of these paper is mainly the quantitative application researchof second-order calibration method and second-order advantage of in complex system,including the chapter2, chapter3and chapter4.Chapter2: The effectiveness of traditional Chinese medicine (TCM) againstvarious diseases urges more low cost, speed and sensitive analytical methods forinvestigating the phamacology of TCM and providing a theoretical basis for clinicaluse. The potential of second-order calibration method was validated for thequantification of the ingredients of schisandra chinensis in complex systems by usingspectrofluorimetry. The results obtained in the present study demonstrated theadvantages of this strategy for multi-target determination in complex matrices.Although the spectra of the analytes are similar and a large number of interferencesalso exist, second-order calibration method could predict the accurate concentrationstogether with reasonable resolution of spectral profiles for analytes of interest owing to’second-order advantage ’. Moreover, the method presented in this work allows one tosimplify experimental procedure as well as reduce the use of harmful chemicalsolvents. Chapter3: In this work, the excitation-emission matrix (EEM) fluorescence dataof guest-host complexes between naproxen enantiomers and β-cyclodextrin was usedto develop second-order calibration method that was subsequently used to determinethe enantiomeric composition of samples of naproxen. The chiral discrimination ofnaproxen enantiomers was realized via their difference in interaction with the chiralcavity of β-cyclodextrin due to their difference in stereochemical structure. Thestrategy combined the use of self-weighted alternating normalized residue fitting(SWANRF) algorithm, for extraction of the pure analyte signal, with the standardaddition strategy, for determination of naproxen enantiomers in presence of matrixeffect caused by the proteins presenting in human urine. The feasible results wereobtained in the molar fraction range from60.0to85.0%of S-naproxen, providingabsolute errors lowers than7.50%. Finally, molecular modeling was performed todetermine the chiral recognition on a molecular level, and the difference in theinteraction energies and the patterns of molecular interactions were discussed. Theresults were in good agreement with the experimental data.Chapter4: This work describes the use of second-order calibration fordevelopment of HPLC-DAD method to quantify nine polyphenols in five kinds ofhoney samples. The sample treatment procedure was simplified effectively relative tothe traditional ways. Baselines drift was also overcome by means of regarding the driftas additional factor(s) as well as the analytes of interest in the mathematical model.The contents of polyphenols obtained by the alternating trilinear decomposition (ATLD)algorithm have been successfully used to distinguish different types of honey. Thismethod shows good linearity (R>0.9900), rapidity (t <7.60min) and accuracy, whichmay be extremely promising as an excellent routine strategy for identification andquantification of polyphenols in the complex matrices.Secondly, the second part of these paper mainly is the study of third-ordercalibration method and third-order advantage, and their applications of quantitativeanalysis in complex system, including the chapter5, chapter6and chapter7.Chapter5: The third-order advantage has been investigated through comparingthe performances of four-way PARAFAC algorithm with those of the PARAFACalgorithm by using simulated and experimental data arrays. All results demonstratedthat third-order calibration could extract more inherent information from the data, andeasily solve the problems associated with serious collinearity and high noise as well asstrong background interference. In addition, the simultaneous determination of threephenolic acids was realized in cosmetic samples. Satisfactory results for the quantification of analytes (average recovery,98.4-101.1%, average relative error,6.3-13.5%) were provided by four-way PARAFAC, evidenced that the third-ordercalibration method could exhibit not only the second-order advantage, but alsoprovided additional advantages.Chapter6: A new third-order calibration algorithm called the alternating cyclicsymmetry quadrilinear decomposition (ACSQLD) was proposed by our group toanalyse the third-order excitation-emission fluorescence data, and the third dimensionwas obtained by recording excitation-emission fluorescence spectra of each sample atseveral pH values. Owing to its unique optimizing approach, ACSQLD is much moreefficient than four-way parallel factor analysis (four-way PARAFAC) in terms of beinginsensitive to excess number of components and fast convergence in calculation. Inorder to check these advantages, simulated and real fluorescenceexcitation-emission-pH data arrays have been processed.Chapter7: In this chapter, the strategy of introducing an extra solvent mode in thethree-way EEMs data was used for the first time in order to investigate the third-orderadvantage through comparing the performances of four-way calibration with those ofPARAFAC. PARAFAC was disappointing in this work due to serious collinearity andhigh background interference. In contrast, the satisfactory results (average recovery,101.5-107.3%, standard deviation,1.4-3.0%) are provided by the third-ordercalibration method based on four-way PARAFAC. In addition, for the predictionDMEM samples, the RMSEP values for schizandrol B obtained by PARAFAC are1.391,2.722,42.89,38.22,35.46g mL-1with five different proportions of ethanol inthe aqueous solvents:40%,50%,60%,70%,80%, respectively, which are manytimes larger than the value (0.0825) obtained by four-way PARAFAC. The resultsdemonstrate that third-order calibration can extract more inherent information from thedata, and can easily solve the problems of serious collinearity and high backgroundinterference. In short, among the advantages of this method over the existing methods,low cost, on-toxic and non-destructive analysis could be cited. |
PDF Full Text Request