Development Of Predictive Models For Diffusion Rate Of Spilled Crude Oil In Niger Delta Soils

ABSTRACT

Simplified predictive numerical models have been developed for diffusion rate of crude oil spilled into Niger Delta soils. The proposed model equations were developed using Fourier transformation, and Green’s function. The induction principles as well as the indicial matrix method of dimensional analysis were adopted in determining the functional parameters that could possibly influence their diffusion rates. These mathematical models correctly predict one – dimensional longitudinal diffusion rate, one-dimensional vertical diffusion rate coefficient, two-dimensional diffusion rate coefficient as well as the concentration distribution in the two spatial directions. The rates of diffusion as functions of volume of crude oil spilled and soil physical properties were correlated. Simple regression techniques were used for the validation of the models with experimental results. The results of the sorption experiment show that the linear isotherm describe the sorption mechanism of the six petroleum samples into the three soil types better with the following distribution coefficient (Kd) values: AF (sand) 0.00861; AF (top) 0.0707; AF (loam) 0.089; UT (sand) 0.1174; UT (top) 0.1324; UT (loam) 0.1352; EG (sand) 0.0697; EG (top) 0.0752; EG (loam) 0.0784; EW (sand) 0.0713; EW (top) 0.0740; EW (loam) 0.909; OBN (sand) 0.1030; OBN (top) 0.1037; OBN (loam) 0.1051; ER (sand) 0.0784; ER (top) 0.0941; ER (loam) 0.1005. The proposed model equations, which are semi-empirical and generalized have, good stability behaviour and are easy to use with the aid of developed simplified and ready to use computational flow schemes.