Chemometrics Analytical Spectroscopy

BACKGROUND: Conventional techniques used for determination of extractable poly(hydroxyalkanoate)s (PHAs) present in the cells, are laborious and destructive. Fourier Transform Infrared (FTIR) spectroscopy is an indirect analytical method to study molecular structures and has been used for the detection and quantification of PHA in intact cells. However, to obtain a fine powder with pure PHA to prepare KBr pellets is a difficult task. The objective of this study was to develop a cheap alternative for quantifying the extractable intracellular PHA, based on a pulverizable solid as a standard material (PEM: Pre Extraction Material) produced by Bacillus megaterium BBST4 strain, instead of pure PHA produced by the same strain. Through statistical analysis the correlation between sample content and its FTIR spectra were determined. The proposed methodology employs typical MID-FTIR equipment and KBr pellets for spectra determination.
RESULTS: Small root-mean-standard errors of prediction of the amount of extractable PHA were obtained using partial least square regression (PLSR1). In order to simplify the determination process, a simple linear regression analysis (SLRA) was also developed, and satisfactory correlations were found.
CONCLUSIONS: Results from PLSR1 and SLRA, calibrated with the proposed standard, indicate that it is a simple, fast and reliable alternative of fitting for determining the extractable PHA in unknown lyophilized cell samples.

Two sensitive, selective and precise stability indicating methods for the determination of isopropamide iodide (ISO), trifluoperazine hydrochloride (TPZ) and trifluoperazine oxidative degradate (DEG) were developed and validated. Method A is a successive derivative ratio spectrophotometric one, which depends on the successive derivative of ratio spectra in two steps using 0.1 N HCl as a solvent and measuring TPZ at 250.4 and 257.2 nm, ISO at 223 and 228 nm and DEG at 210.6, 213 and 270.2 nm. Method B is mean centering of ratio spectra which depends on using the mean centered ratio spectra in two successive steps and measuring the mean centered values of the second ratio spectra at 322, 355 and 339 nm for TPZ, ISO and DEG, respectively. Factors affecting the developed methods were studied and optimized, moreover, they have been validated as per ICH guidelines and the results demonstrated that the suggested methods are reliable, reproducible and suitable for routine use with short analysis time. Statistical analysis of the two developed methods with the reported one using F- and Student’s t-test showed no significant difference regarding accuracy and precision.

Chemometric assisted spectrophotometry and HPLC methods have been developed for the
simultaneous determination of Amlodipine besylate and Telmisartan. The two chemometric
methods i.e. Inverse least square (ILS) and Classical least square (CLS) methods were
successfully applied to quantify each drug in the mixture using the information included in
the UV absorption spectra of appropriate solutions in the range of 230-302 nm with the
intervals of 4 nm at 19 wavelengths. The HPLC method was developed and the three
methods (ILS, CLS and HPLC) were successfully applied to marketed formulation and the
results were compared statistically which showed no significant difference among the results.

Mathematical Processing of  Spectral Data in Analytical  Chemistry: A Guide to Error Analysis

By Joseph Dubrovkin

Cambridge Scholars Publishing

Release Date: Aug. 1, 2018

This book will appeal to both practitioners and researchers in industrial and university analytical laboratories, as well as students specializing in analytical spectroscopy and chemometrics. The subjects covered include the advanced principles of calibration (univariate and multivariate) and the estimation of the peak parameters in spectra with overlapping components. This book differs from existing studies on the subject in that it provides easily reproducible computer calculations illustrating its significant theoretical statements. As such, it can also serve as a practical guide to lecturers in analytical spectrometry and chemometrics.

ABOUT THE AUTHOR

Joseph Dubrovkin gained a degree in Automatics from the Aviation Institute, Russia, in 1968, and Doctoral degrees in Technical Sciences and Physics and Mathematics from Leningrad State University, Russia, in 1979 and 1989 respectively. He was a Lecturer at the Aviation Institute and the Pharmaceutical Institute, Russia, and Western Galilee College (department of Bar-Ilan University), Israel, before retiring.

20% discount on online orders using the discount code ERROR20 – purchase through www.cambridgescholars.com or email orders@cambridgescholars.com

This book will appeal to both practitioners and researchers in industrial and university analytical laboratories, as well as students specializing in analytical spectroscopy and chemometrics. The subjects covered include the advanced principles of calibration (univariate and multivariate) and the estimation of the peak parameters in spectra with overlapping components. This book differs from existing studies on the subject in that it provides easily reproducible computer calculations illustrating its significant theoretical statements. As such, it can also serve as a practical guide to lecturers in analytical spectrometry and chemometrics.

The aim of this work is to optimize a spectrofluorimetric method for the determination of cefdinir (CFN) using the Taguchi method. The proposed method is based on the oxidative coupling reaction of CFN and cerium(IV) sulfate. The quenching effect of CFN on the fluorescence of the produced cerous ions is measured at an emission wavelength (λ em) of 358 nm after exci-tation (λ ex) at 301 nm. The Taguchi orthogonal array L 9 (3 4) was designed to determine the optimum reaction conditions. The results were analyzed using the signal-to-noise (S/N) ratio and analysis of variance (ANOVA). The optimal experimental conditions obtained from this study were 1 mL of 0.2% MBTH, 0.4 mL of 0.25% Ce(IV), a reaction time of 10 min and methanol as the diluting solvent. The calibration plot displayed a good linear relationship over a range of 0.5–10.0 μg/mL. The proposed method was successfully applied to the determination of CFN in bulk powder and pharmaceutical dosage forms. The results are in good agreement with those obtained using the comparison method. Finally, the Taguchi method provided a systematic and efficient methodology for this optimization, with considerably less effort than would be required for other optimizations techniques.

Authors: Višňák, J.; Steudtner, R.; Kassahun, A.; Hoth, N. Natural waters’ uranium level monitoring is of great importance for health and environmental protection. One possible detection method is the Time-Resolved Laser-Induced Fluorescence Spectroscopy (TRLFS), which offers the possibility to distinguish different uranium species. The analytical identification of aqueous uranium species in natural water samples is of distinct importance since individual species differ significantly in sorption properties and mobility in the environment. Samples originate from former uranium mine sites and have been provided by Wismut GmbH, Germany. They have been characterized by total elemental concentrations and TRLFS spectra. Uranium in the samples is supposed to be in form of uranyl(VI) complexes mostly with carbonate (CO3(2-)) and bicarbonate (HCO3(-)) and to a lesser extent with sulphate (SO4(2-)), arsenate (AsO4(3-)), hydroxo (OH(-)), nitrate (NO3(-)) and other ligands. Presence of alkaline...

Authors: Višňák, J.; Steudtner, R.; Kassahun, A.; Hoth, N.
Natural waters’ uranium level monitoring is of great importance for health and environmental protection. One possible detection method is the Time-Resolved Laser-Induced Fluorescence Spectroscopy (TRLFS), which offers the possibility to distinguish different uranium species. The analytical identification of aqueous uranium species in natural water samples is of distinct importance since individual species differ significantly in sorption properties and mobility in the environment. Samples originate from former uranium mine sites and have been provided by Wismut GmbH, Germany. They have been characterized by total elemental concentrations and TRLFS spectra. Uranium in the samples is supposed to be in form of uranyl(VI) complexes mostly with carbonate (CO3(2-)) and bicarbonate (HCO3(-)) and to a lesser extent with sulphate (SO4(2-)), arsenate (AsO4(3-)), hydroxo (OH(-)), nitrate (NO3(-)) and other ligands. Presence of alkaline earth metal dications (M = Ca(2+), Mg(2+), Sr(2+)) will cause most of uranyl to prefer ternary complex species, e.g. Mn(UO2)(CO3)3(2n-4) (n = 1, 2). From species quenching the luminescence, Cl- and Fe2+ should be mentioned. Measurement has been done under cryogenic conditions to increase the luminescence signal. Data analysis has been based on Singular Value Decomposition and monoexponential fit of corresponding loadings (for separate TRLFS spectra, the “Factor analysis of Time Series” (FATS) method) and Parallel Factor Analysis (PARAFAC, all data analysed simultaneously). From individual component spectra, excitation energies T_{00}, uranyl symmetric mode vibrational frequencies omega_{gs} and excitation driven U-Oyl bond elongation DeltaR have been determined and compared with quasirelativistic (TD)DFT/B3LYP theoretical predictions to cross-check experimental data interpretation.

Working on the Global NIR ID (Identification) Project, the author needed a way to create a user-friendly, straightforward, easy-to-follow, procedurally driven software application that could be used to perform the functions of his role in the development, validation, and maintenance of the FT-NIR material discriminators at Perrigo (Allegan, MI). With the company receiving well over 500 raw materials approved for use, it had to find a way to identify raw materials in an efficient manner that was equivalent to or exceeded the quality of current compendial identification methods. NIR ID was selected using the ABB Bomem (Quebec, Canada) MB-160 series (now FTLA-2000-160 series) and the Advanced Infra Red (AIRS) software (DHC Analysis Inc., Cleveland, OH).

Individual discriminate-partial least squares (D-PLS) models were employed to reduce or eliminate the probability of a material falsely passing (beta-error) as another material. Furthermore, the confirmation configuration selection aspect of the AIRS software allowed for an easy-to-assess mode of failing in case a multiple ID or beta-error were to occur.

Fourier transform infrared (IR) spectroscopy in combination with multivariate data analysis is a versatile tool that can be applied to disease diagnosis. However, a rigorous validation of the obtained models is necessary in order to obtain robust results. This work evaluates the advantages of the use of permutation testing for determining the statistical significance if the misclassification errors obtained from IR based diagnostic
models through cross validation (CV). The model performance, estimated by CV, is compared to a distribution of CV-performance values obtained using randomly permuted class labels. The distribution of ‘random CV-values’ is considered as a null distribution and used to establish the significance of the
model estimators obtained using real class labels. ATR-FTIR spectra of serum samples were classified using random forest (RF) classifiers according to two criteria, the tag number (a randomly assigned pseudo class membership) and the level of urea (real class). CV errors obtained were compared to the
null distribution of CV errors from a permutation test and an independent validation set. The procedure was evaluated testing typical conditions leading to overoptimistic estimations provided by the CV like e.g. the size of subsamples used during CV, variable selection and the use of replicates. Results show that for the tag number (pseudo class), CV indicated classification errors between 23 and 33% depending
on the subsample size employed. Those values were even lower when variable selection or replicates were used. However, permutation testing indicated that those CV errors were non-significant. In contrast, for sample classification according to their levels of urea, all cross validation errors were found to be significant. Although the proposed method is computationally intensive, it provides a simple way of calculating an empirical p-value of the CV-estimator, thus establishing the statistical significance and providing a feasibility indicator especially useful for studies where the number of samples is limited.

Three methods are presented for the simultaneous determination of diloxanide furoate (DLX) and metronidazole (MTR), used for their antiprotozoal and antiamoebic effect, in the presence of DLX alkaline degradates and in pharmaceutical formulations, without previous separation. The first method is chemometric-assisted spectrophotometry, in which principal component regression and partial least squares were applied. These two approaches were successfully applied to quantify each drug in the mixture using the information included in, the absorption spectra in the range of 225-320 nm. The second method is TLC-densitometry, in which the binary mixture and degradates were separated on silica gel plates using a chloroform-acetone-glacial acetic acid (9.5 + 0.5 + 0.07, v/v/v) mobile phase and the bands were scanned at 254 nm. The last method is HPLC, in which DLX, MTR, and degradates were separated using the mobile phase acetonitrile-0.05 M dibasic potassium phosphate (25 + 75, v/v), adjusted to pH 4 with orthophosphoric acid, at a flow rate of 1 mL/min, on a C18 analytical column. Detection was at 254 nm. The proposed methods were successfully applied for the analysis of DLX and MTR in pharmaceutical formulations, and the results were statistically compared with a reported spectrophotometric method.

Pharmaceutical counterfeiting is a life threatening problem affecting all countries. Counterfeit medicines may be encountered anywhere in conventional markets or from the Internet. This paper proposes a rapid and non-destructive near-infrared spectroscopic method for the identification of counterfeit medicines using the minimum number of authentic samples. As little as twenty spectra from ten tablets from a batch are required to compare a test sample to its authentic counterpart. In this respect, tablets are measured as received and the correlation coefficient of the SNV-D2 spectra between the authentic sample and the test sample is determined. A correlation coefficient of lower than 0.95 indicates that the batch fails identification. In this case, if enough authentic samples are available, principal component analysis (PCA) could be applied. The PCA scores plot of the authentic and counterfeit samples with the 95% equal frequency ellipses drawn around the authentic sample set is effective in identifying counterfeits. The method could identify 82 known counterfeit medicines out of 201 medicines supplied from the Internet and the World market. However, it is still a comparative method to identify potential counterfeits and cannot identify products without authentic samples.