ANN Prediction of Equilibrium Constants in Aqueous Solutions

Investor logo

Warning

This publication doesn't include Faculty of Economics and Administration. It includes Faculty of Science. Official publication website can be found on muni.cz.
Authors

LUBAL Přemysl HAVEL Josef FARKOVÁ Marta

Year of publication 2002
Type Article in Proceedings
Conference Book of Abstracts of International Chemometric Conference CHEMOMETRICS VI
MU Faculty or unit

Faculty of Science

Citation
Field Analytic chemistry
Keywords artificial neural networks; equilibrium constants
Description The knowledge of stability constants is important in all branches of chemistry, chemical technology, environment, etc. The equilibrium (formation, stability) constants in analytical chemistry are used in order to understand speciation in development of analytical procedures or in the environment. The measurement of large numbers of equilibrium constants of different reactions varying experimental conditions (ionic strength, temperature, etc.) is not attractive option. Therefore accurate and reliable methods for determination of equilibrium constants are desirable. In practice, different equations are used for prediction of equilibrium constants for given experimental conditions (ionic strength, temperature, etc.). The precision of prediction is dependent on the number of experimental points and relationship applied in the fitting procedure. Recently we proposed the application of artificial neural networks (ANN's) for evaluation of equilibrium constants from experimental data obtained by means of different experimental techniques. In this contribution, the method of equilibrium constants prediction for different ionic strengths and temperatures using "soft" modelling with ANN's was examined and compared with results obtained by "hard" modelling. This proposed methodology allows rapidly and with sufficient accuracy to predict formation constants for given experimental conditions. The results are independent on the model and also are not sensitive to error of formation constants. This alternative model-free approach for prediction of stability constants can be used in practice.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.