Šime Ukić

Application of Different Artificial Neural Networks Retention Models for Multi-Criteria Decision-Making Optimization in Gradient Ion Chromatography

In this work, the principles of multi-criteria decision-making were used to develop an efficient optimization strategy in gradient elution ion chromatographic analysis. Two different artificial neural network retention models (multi-layer perceptron and radial basis function), three different separation criterion functions (chromatography response function, separation factor product and normalized retention difference product), and four different robustness criterion functions (CR1-CR4) were examined. The shape of the calculated separation vs the robustness response surface was used as principal criterion. Analysis time and minimum separation of adjacent peaks were additional criteria. The results showed that the radial basis artificial neural network retention model in combination with normalized retention difference product separation criterion function and CR3 robustness criterion function provided the optimal gradient ion chromatographic analysis.

Application of a Gradient Retention Model Developed by Using Isocratic Data for the Prediction of Retention, Resolution, and Peak Asymmetry in Ion Chromatography

Abstract In this work a model was developed for the prediction of retention time, resolution, and peak asymmetry in gradient elution mode by using isocratic experimental data. The predictive performance and generalization ability of the developed model was extensively tested by using an external experimental data set. The analysis of errors was performed in order to discuss and explain characteristics of the model. It was shown that the model performed satisfactorily and that it could be used for a modeling procedure in the optimization part of the ion chromatography method development.

Evaluation of separation in gradient elution ion chromatography by combining several retention models and objective functions.

In this work, three different methods for modeling of gradient retention were combined with several optimization objective functions in order to find the most appropriate combination to be applied in ion chromatography method development. The system studied was a set of seven inorganic anions (fluoride, chloride, nitrite, sulfate, bromide, nitrate, and phosphate) with a KOH eluent. The retention modeling methods tested were multilayer perceptron artificial neural network (MLP-ANN), radial-basis function artificial neural network (RBF-ANN), and retention model based on transfer of data from isocratic to gradient elution mode. It was shown that MLP retention model in combination with the objective function based on normalized retention difference product was the most adequate tool for optimization purposes.

Prediction of diesel fuel cold properties using artificial neural networks

In this paper, two neural networks, multilayer perceptron and networks with radial-basis function, were used to predict important cold properties of commercial diesel fuels, namely cloud point and cold filter plugging point. The developed models predict the named properties using cetane number, density, viscosity, contents of total aromatics, and distillation temperatures at 10, 50, and 90 vol. % recovery as input data. The training algorithms, number of hidden layer neurons, and number of training data points were optimized in order to obtain a model with optimal predictive ability. The results indicated better prediction of cloud and cold filter plugging points in the case of multilayer perceptron networks. The obtained absolute error mean for the optimal neural network models (0.58°C for the cloud point and 1.46°C for the cold filter plugging point) are within the range of repeatability of standard cold properties determination methods.

Prediction of the chromatographic signal in gradient elution ion chromatography

This study describes the development of a signal prediction model in gradient elution ion chromatography. The proposed model is based on a retention model and generalized logistic peak shape function which guarantees simplicity of the model and its easy implementation in method development process. Extensive analysis of the model predictive ability has been performed for ion chromatographic determination of bromate, nitrite, bromide, iodide, and perchlorate, using KOH solutions as eluent. The developed model shows good predictive ability (average relative error of gradient predictions 1.94%). The developed model offers short calculation times as well as low experimental effort (only nine isocratic runs are used for modeling).

Optimization of IC Separation Based on Isocratic-to-Gradient Retention Modeling in Combination with Sequential Searching or Evolutionary Algorithm.

Gradient ion chromatography was used for the separation of eight sugars: arabitol, cellobiose, fructose, fucose, lactulose, melibiose, N-acetyl-D-glucosamine, and raffinose. The separation method was optimized using a combination of simplex or genetic algorithm with the isocratic-to-gradient retention modeling. Both the simplex and genetic algorithms provided well separated chromatograms in a similar analysis time. However, the simplex methodology showed severe drawbacks when dealing with local minima. Thus the genetic algorithm methodology proved as a method of choice for gradient optimization in this case. All the calculated/predicted chromatograms were compared with the real sample data, showing more than a satisfactory agreement.

Novel criteria for fast searching for optimal method in gradient ion chromatography: an integrated approach.

In this article, an integrated approach for prediction and optimization in ion chromatography (IC) was presented. The approach provides a fast and reliable insight in the elution behavior of an IC system. The predictions are based on a mathematical model that predicts ion retentions (for both isocratic and gradient modes) by using an empirical isocratic model. Other chromatographic values significant for the optimal elution conditions (resolution, peak asymmetry) are calculated quickly and easily from the predicted retention values of characteristic points of a chromatographic peak. Every day, IC users might find this approach a suitable tool for finding optimal IC elution conditions in a given system.

PREDICTION OF NONLINEAR GRADIENT SIGNAL IN ION CHROMATOGRAPHY BASED ON “EXPERIMENT-FREE” METHODOLOGY

One of the strategies that might be applied in ion chromatographic (IC) multi-segment gradient optimization is the simplification of gradient profile and its approximation by nonlinear elution. This work describes the IC signal modeling strategy for linear and nonlinear gradients. The performance characteristics of developed models do not depend on the linearity of gradient profile, but rather on initial gradient of eluent competing ion concentration. Four different IC models were developed and tested, requiring three, two, or single experimental data points. Moreover, a model based on “experiment-free” methodology using standard column manufacturer quality assurance documents was developed. It is shown that all the developed models have good predictive ability that decreases by reducing the model complexity and number of experimental data required. However, even the “experiment-free” methodology offers reasonable performance characteristics (average of relative error 4.99%) and might be recommended for IC method development purposes.

Influence of process parameters on the effectiveness of photooxidative treatment of pharmaceuticals

ABSTRACT In this study, UV-C/H2O2 and UV-C/ processes as photooxidative Advanced oxidation processes were applied for the treatment of seven pharmaceuticals, either already included in the Directive 2013/39/EU “watch list” (17α- ethynylestradiol, 17β-estradiol) or with potential to be added in the near future due to environmental properties and increasing consumption (azithromycin, carbamazepine, dexamethasone, erythromycin and oxytetracycline). The influence of process parameters (pH, oxidant concentration and type) on the pharmaceuticals degradation was studied through employed response surface modelling approach. It was established that degradation obeys first-order kinetic regime regardless structural differences and over entire range of studied process parameters. The results revealed that the effectiveness of UV-C/H2O2 process is highly dependent on both initial pH and oxidant concentration. It was found that UV-C/ process, exhibiting several times faster degradation of studied pharmaceuticals, is less sensitive to pH changes providing practical benefit to its utilization. The influence of water matrix on degradation kinetics of studied pharmaceuticals was studied through natural organic matter effects on single component and mixture systems.

Photooxidative Degradation of Pesticides in Water; Response Surface Modeling Approach

Abstract The conventional water treatment technologies, mostly relying on physical and biological processes, seem to be inadequate for effective removal of priority substances such as pesticides, while advanced oxidation processes showed a good performance for the same purpose. The aim of the study was to evaluate UV-C/H2O2 and UV-C/S2O82– processes for treating seven pesticides listed as priority substances (PS-Ps) within the EU Water Framework directive; alachlor, atrazine, chlorfenvinphos, cybutryne, diuron, isoproturon and simazine. The influence of pH and pollutant/oxidant ratio ([PS-P]:[OX]) on the degradation kinetics was evaluated using full factorial plan and response surface modeling. The degradation of PS-Ps by both photooxidation processes obeyed first-order kinetics. Degradation kinetics of PS-Ps is highly depended on [PS-P]: [OX], while pH has minor significance, particularly in UV-C /S2O82– process. In most cases degradation kinetics by UV-C /S2O82– was several times faster in comparison to UV-C /H2O2 process. In addition, the inhibitory influence of NOM on the degradation of PSs in the mixture was determined.