In order to reduce potential hazards to the environment resulting from production and use of chemicals, some governments have enacted laws and corresponding regulations. Thus, for instance, the German “law on hazardous substances,” and the regulation regarding a.o. submission of test data on physicochemical properties of new substances, have been ratified by the Federal Republic of Germany legislation in 1980 and 1981, respectively. This implies that before production, marketing, and use of a new chemical, its potential hazard to human beings and the environment has to be checked. Therefore it is necessary to have available accurate information about the probable fate of this new chemical in the environment, i.e., its distribution among the different environmental compartments. To this end, the submission of basic physicochemical data of a new chemical, such as boiling point, vapor pressure, water solubility, fat solubility, and partition coefficient in the system n-octanol/water, is mandatory for its registration. These data are meant to be used by the official evaluating department as a basis for describing the distribution behavior of the substance in the environment. However, there is no generally accepted procedure yet which would allow the reliable prediction. Such a calculation procedure should comprise all factors decisive for distribution and degradation of chemicals in the environment, according to their significance. We have taken a first step in this direction by carrying out the following experiments, with the aim of developing such a distribution model as well as identifying the physicochemical properties of chemicals required for the prediction of distribution patterns. The distribution patterns of 12 selected substances, having significant differences with respect to their physicochemical properties, were determined via radiotracer technique, in a defined, standardized, terrestrial ecosystem. The results of the experiments were compared with one another as well as with values calculated by use of mathematical models from the literature. These comparisons show that there are certain correlations between the distribution patterns obtained by experiments and those calculated by the stationary distribution models. However, these also indicate that some of the experimental findings cannot be explained within the framework of stationary distribution models. Kinetic models represent a more universal approach toward predicting the distribution of chemicals in an ecological system. Various prototypes are described in the literature but we could not use them due to their mathematical shortcomings. Therefore, we developed a kinetic model which provides the prediction formulas representing the probable distribution of chemicals in a standardized, terrestrial ecosystem, consisting of the compartments soil, plants, and air. The observed distribution patterns of the substances tested by us can be represented by two additive terms. The first one is defined by the conditions existing at the start of the distribution experiment in the standard ecosystem. The second term is specified by the physicochemical properties influencing the time course of the distribution process of the substance. These model parameters (diffusion, partition, and mass transfer coefficients as well as the velocity constants of the degradation processes) were adjusted to the data obtained from distribution experiments. The numerical values of the model parameters correlate satisfactorily with particular physicochemical characteristics of the substances while the model equations reproduce the experimentally determined distribution patterns as well as their time slope property. In summary, the distribution pattern of a chemical in the standardized terrestrial ecosystem can be predicted by mathematical calculation, if in addition to the parameters of the particular ecological system, the physicochemical characteristics of the substance, particularly Ostwald's solubility, soil/water partition coefficient, and kinetic constants of degradation processes are known. Additionally it is shown that there is the possibility of generalizing the kinetic model for the purpose of predicting distribution patterns of chemicals in nonstandardized ecological systems.
Read full abstract