Josef Havel

Predicting sorption of pharmaceuticals and personal care products onto soil and digested sludge using artificial neural networks

A comprehensive analytical investigation of the sorption behaviour of a large selection of over-the-counter, prescribed pharmaceuticals and illicit drugs to agricultural soils and freeze-dried digested sludges is presented. Batch sorption experiments were carried out to identify which compounds could potentially concentrate in soils as a result of biosolid enrichment. Analysis of aqueous samples was carried out directly using liquid chromatography-tandem mass spectrometry (LC-MS/MS). For solids analysis, combined pressurised liquid extraction and solid phase extraction methods were used prior to LC-MS/MS. Solid-water distribution coefficients (K(d)) were calculated based on slopes of sorption isotherms over a defined concentration range. Molecular descriptors such as log P, pK(a), molar refractivity, aromatic ratio, hydrophilic factor and topological surface area were collected for all solutes and, along with generated K(d) data, were incorporated as a training set within a developed artificial neural network to predict K(d) for all solutes within both sample types. Therefore, this work represents a novel approach using combined and cross-validated analytical and computational techniques to confidently study sorption modes within the environment. The logarithm plots of predicted versus experimentally determined K(d) are presented which showed excellent correlation (R(2) > 0.88), highlighting that artificial neural networks could be used as a predictive tool for this application. To evaluate the developed model, it was used to predict K(d) for meclofenamic acid, mefenamic acid, ibuprofen and furosemide and subsequently compared to experimentally determined values in soil. Ratios of experimental/predicted K(d) values were found to be 1.00, 1.00, 1.75 and 1.65, respectively.

Neural networks for optimization of high-performance capillary zone electrophoresis methods. A new method using a combination of experimental design and artificial neural networks

Artificial neural networks (ANN) offer attractive possibilities for providing non-linear modeling of response surfaces and optimization in capillary zone electrophoresis (CZE) when the underlying mechanisms are very complex or not well known or understood, in comparison with (non)-linear regression methods. The application of ANN in optimization of CZE methods has been examined and a new method, based on the combination of experimental design and ANN methods, which offers considerable effectiveness, has been developed.

Multiparametric curve fitting--X: a structural classification of programs for analysing multicomponent spectra and their use in equilibrium-model determination.

A functional structure-classification of programs for analysis of spectra elucidates their efficiency for determination of the stoichiometric indices, stability constants and molar absorptivities of complex species. SQUAD (84) introduces new functional units for (i) determination of the number of light-absorbing species, (ii) a rigorous fitness test, (iii) plotting three-dimensional graphs of a paraboloid minimum response-surface as a function of two selected parameters, and a graph of the fitted absorbance response-plane, (iv) simultaneous estimation of stoichiometric indices and stability constants, (v) simulation of an absorbance matrix data by loading with random errors related to the instrumental variance of the absorbance. A guide to experimental procedure and computational strategy for chemical model determination is given and nine diagnostic tools useful in finding the number of species present and their stoichiometry and stability constants by regression analysis of spectra are tested, by use of literature data.

Prediction of retention times for anions in ion chromatography using Artificial Neural Networks

SummaryAn Artificial Neural Network (ANN) was investigated as a method to model retention times of anions in nonsuppressed and suppressed ion chromatography (IC) using a range of eluents and stationary phases, with the results being compared to those obtained using mathematical retention models. The optimal ANN architecture was determined for six specific IC cases of increasing complexity. Analysis of the retention times predicted using the ANN and those predicted by the mathematical models showed that the ANN approach yielded superior performance in all of the above cases. The use of a limited training data set configured in a central composite experimental design was suitable for application of the ANN to non-suppressed IC but was not applicable to suppressed IC, for which a more extensive training data set was necessary.

The acidobasic and complexation properties of humic acids: Study of complexation of Czech humic acids with metal ions

The acid-base and complexation properties of humic acids (HAs) extracted from bohemian brown coals were studied. The acid-base behavior corresponds with the model of HA as a mixture of mono- and diprotic acids. This model was also verified on commercial HA substances (Aldrich and Fluka). HA binds strongly with heavy metal ions and the highest stability constant of HA-metal ion complexes was observed for copper(II). Stability constant values were found to decrease in the order: Cu(2+)>Ba(2+)>Pb(2+)>Cd(2+)>Ca(2+). Both acidobasic models for HA alone and those for HA-metal ion interactions were proposed and the computational methodology for polyelectrolyte equilibria studies demonstrated.

Prediction of retention times for anions in linear gradient elution ion chromatography with hydroxide eluents using artificial neural networks.

The feasibility of using an artificial neural network (ANN) to predict the retention times of anions when eluted from a Dionex AS11 column with linear hydroxide gradients of varying slope was investigated. The purpose of this study was to determine whether an ANN could be used as the basis of a computer-assisted optimisation method for the selection of optimal gradient conditions for anion separations. Using an ANN with a (1, 10, 19) architecture and a training set comprising retention data obtained with three gradient slopes (1.67, 2.50 and 4.00 mM/min) between starting and finishing conditions of 0.5 and 40.0 mM hydroxide, respectively, retention times for 19 analyte anions were predicted for four different gradient slopes. Predicted and experimental retention times for 133 data points agreed to within 0.08 min and percentage normalised differences between the predicted and experimental data averaged 0.29% with a standard deviation of 0.29%. ANNs appear to be a rapid and accurate method for predicting retention times in ion chromatography using linear hydroxide gradients.

Laser desorption ionization of red phosphorus clusters and their use for mass calibration in time‐of‐flight mass spectrometry

Phosphorus clusters P(n) (n = 1-89) are easily formed from red phosphorus by laser desorption ionization (LDI) and they cover a range of up to approx. m/z 3000 in both positive and negative ion mode. The clusters are singly charged and the spectra are simple because phosphorus is monoisotopic. The mass spectra can be measured with an acceptable resolution and intensity. The use of positively charged P(n) clusters for calibration in mass spectrometry was examined and it was demonstrated that in external calibration a standard deviation of +/-0.04 m/z units can be achieved even when using a common commercial matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) instrument. When used as internal standards the P(n) clusters react with some analytes - C(60) and C(70) fullerenes and cucurbituril[8], for example. It was also found that red phosphorus is a suitable MALDI matrix for peptides and proteins, illustrated by the examples of a Calmix mixture of bradykinin, angiotensin, renin, adrenocorticotropic hormone ACTH fragment 18-359 and insulin, and of insulin alone.

Separation of polyphenols in Canary Islands wine by capillary zone electrophoresis without preconcentration

A method for separation and determination of polyphenols in wine by capillary zone electrophoresis (CZE) without any preconcentration step is described. The sensitivity and limits of detection for gentisic and p-coumaric acid are better than those previously published. The effect of a possible C18 solid-phase extraction prior to the CZE analysis was examined. The developed optimized method (without any extraction step) was applied to the analysis of wines from Tenerife, Canary Islands.

Evaluation of multicomponent kinetic analysis data by a partial least squares calibration method

A partial least squares (PLS) calibration method is suggested for the evaluation of multicomponent kinetic data. The advantage is that no kinetic model is assumed and only the information from the kinetic curves themselves is used. In the procedure proposed, a model is developed from samples of known composition, i.e., a calibration set, and is then used to predict unknowns, i.e., concentrations of the components in a test set. The use of the PLS method for multicomponent kinetic analysis is demonstrated on several simulated and experimental data. It was shown that it is possible to determine analytes in mixtures without any previous knowledge either of the rate constants, mechanism of the reactions or molar absorptivities of the components. The method has been validated on several two- and three-component simulated data and applied to the simultaneous kinetic determination of Co and V and the stopped-flow flow injection determination of Zn, Co and Fe.

Capillary zone electrophoresis for the separation and characterization of humic acids

Abstract Capillary zone electrophoresis (CZE) was used for the separation and characterization of humic acids (HAs) of different origin. UV–Vis detection and/or fast scanning of spectra during the separation was applied for the detection. From several background electrolytes (BGEs) studied, it was found that especially those containing amino acids are the most suitable and led to the separation of HAs into several fractions. The composition and pH of buffers were optimized. Up to 30 fractions were separated using a high boric acid concentration (350 m M ) BGE. The results can be used as real “fingerprints” to characterize HAs of different origin.