Files and links (1)
PDFOpen Access (License Unspecified), Open Access
Dissertation
Prediction of aqueous solubility and Henry's constant from molecular structure
Doctor of Philosophy (Ph.D.), Drexel University
1988
DOI:
https://doi.org/10.17918/00010235
Abstract
Aqueous Solubility and Henry's Constant data from the literature were compiled for a wide range of chemicals of environmental concern. Using a semi-theoretical approach, the solute-solvent interactions were modeled by Quantitative Structure Activity Relationship (QSAR) techniques to derive predictive equations for the two properties, without reliance on any experimental inputs. Molecular connectivity indices and a polarizability parameter were used to model the cavity formation and the dipole-dipole phenomena in solute-solvent interactions. A training data set of 145 mixed class of solid and liquid solutes was used to derive a general model for aqueous solubility with an r of 0.970 and standard error of 0.311. The predictive ability of the model was demonstrated using a testing set of 55 new chemicals. Appropriate statistical tests were used to demonstrate the integrity of the general model. Finally, the model was shown to fit the data for an expanded data set of 315 compounds with an r of 0.974 and standard error of 0.291. In the study of Henry's Constant, a data set of 180 chemicals was used to derive a preliminary model with an r of 0.965 and standard error of 0.445. By introducing a simple indicator variable, this model was refined to fit the data more precisely with an r of 0.99 and standard error of 0.262. The robustness of this model too was demonstrated, as well as its predictive ability, by applying it on a testing data set of 20 miscellaneous chemicals. The maximum error on these 20 chemicals was -0.63 log units, and for 10 of them, the error was within +/-0.2 log units. The connectivity-polarizability approach developed in this study was shown to be a simple and powerful one to model solute-solvent interactions in partitioning phenomena. The primary advantage of this approach is the simple to formulate descriptors, calculable by rigid and consistent algorithms for all classes of compounds from their molecular structure, without any experimental inputs. To make this approach more useful and to demonstrate the utility of QSAR techniques, the entire study was carried out in a personal computer environment.
Metrics
22 File views/ downloads
20 Record Views
Details
- Title
- Prediction of aqueous solubility and Henry's constant from molecular structure
- Creators
- Nagamany Nirmalakhandan
- Contributors
- Richard E. Speece (Advisor) - Drexel University, Drexel University (1970-)
- Awarding Institution
- Drexel University
- Degree Awarded
- Doctor of Philosophy (Ph.D.)
- Publisher
- Drexel University; Philadelphia, Pennsylvania
- Number of pages
- ix, 132 pages
- Resource Type
- Dissertation
- Language
- English
- Academic Unit
- College of Engineering (1970-2026); Drexel University
- Other Identifier
- 991021888985904721