CAPILLARY ELECTROPHORESIS CHIRAL SEPARATION MODELLING WITH THE USE OF ARTIFICIAL NEURAL NETWORKS
| Authors | |
|---|---|
| Year of publication | 2002 |
| Type | Article in Proceedings |
| Conference | CHIRANAL 2002 |
| MU Faculty or unit | |
| Citation | |
| Field | Analytic chemistry |
| Keywords | artificial neural networks; capillary electrophoresis; chiral separation |
| Description | Recent development and future trends of enantioseparations in capillary electrophoresis have been reviewed by Chankvetadze et al. On the base of exact physicochemical description using e.g. CELET program the stability constants of either chiral or non-chiral inclusion complexes can be calculated. As for review we refer to Vespalec et al. Recently, we have shown that "soft" modelling of achiral CE separation processes is possible using a combination of artificial neural networks (ANN) and experimental design. Possibility of enantiomers quantification from unresolved peaks was also demonstrated. In this work we are examining possibility of chiral separation "soft" modelling with ANN. It was found that, using suitable ANN architecture, the description of chiral separation is possible with sufficient accuracy. The advantage is that it is not necessary to know or determine chiral selector - enantiomers stability constants and/or the separation mechanism. Using combination of suitable experimental design and ANN architecture, the prediction of optimal conditions for the separation of enantiomers is possible. |
| Related projects: |