Özet:
The main purpose of this work is to develop a mathematical model based on the Valance Molecular Connectivity theory which is initiated as an extension of graph theory having its roots especially in structure-activity studies concerning drug-design and toxicity prediction in pharmaceutical chemistry, to evaluate the compartmental distribution of various aromatic pollutants in the environment. During the last decade concerns have been expressed as to the "best way" to assess the potential hazards posed by exposure to chemical substances. In response to these concerns, the chemical group of OECD initiated a hazard assessment project to examine the available methods for hazard assessment of chemicals. The main purpose being to determine how information on theultimate fate and effects of a chemical can be derived from the set of premarket data, the group accepted four models for the estimation of exposure potentials of chemicals within environmental compartments of major concern. Ail of the four models accepted by OECD chemical group was based on the concept of fugacity and essentially needed the following data e molecular weight e water solubility e vapor pressure e soil sorption constant e octanol-water partition coefficient However, the availability and precission of the set of data proposed, especially for chemicals which have rather large and complicated structures was a drawback of these models. Hence, a reasonable model to evaluate this distribution quantitatively without needing such hard-to-gather data would have been a remarkable advance in prediction of environmental hazards that are going to be caused by unknown chemicals irrespective of size and conformation. In this work a new mathematical model (VMCI) based only on the topological characteristics of molecules is deveioped in order to evaluate the distribution of chemicals within various compartments of the environment such as air, soil, water, biota, suspended solids and sediment. The Valance Molecular Connectivity Index, which is the basis of this new model is known to correlate significantly with a number of structure dependent physicochemical properties, and by this wock lt has now been shown to have a high degree of correlation also with water solubilities, vapor pressures and partition coefficients hence wlth the partitioning properties of molecules. As a result, by comparing the two completely different methods of evaluation (VMCI and Level I Fugacity Models) this work proved that it is possible to predict the compartmental distribution of any aromatic compound quantitatively within a high accuracy just by considering the molecular geometry. The superiority of this model compared to fugacity dependent models accepted by OECD lies in the fact that it does not require the physical data such as solubility, vapor pressure or partition coefficient. Relevantly another superiority comes out to be the ability to stay as accurate when even very complicated molecules are investigated while for fugacity dependent models accuracy diminishes due the difficulties in obtaining the experimental data.
