Jerker Jarsjö

Retention and volatilisation of kerosene: Laboratory experiments on glacial and post-glacial soils

Abstract The influence of environmental conditions and soil characteristics on the retention and volatilisation of kerosene hydrocarbons in soils is investigated through laboratory experiments in six different glacial and post-glacial soils. The soils ranged between 0.55 and 1.80 g cm −3 in density, 29% and 70% in porosity, 0.4% and 28% in organic matter content, and 0% and 51% in clay content. The water retention capacity (WRC) could be estimated on basis of the bulk density and the sand, silt, clay and organic matter contents for many of the soils; for these soils a simple linear relation was identified between WRC and the kerosene retention capacity (KRC). Furthermore, the combined effects of soil porosity and soil moisture content on the KRC were found to be significant and could be quantified by a linear relationship. For moisture contents that can be expected in humid climate zones, the KRC will be relatively low and exhibit small variations between different soils. The effect of temperature on KRC was found to be small. The kerosene volatilisation flux was quantified as a function of time by a power law relationship, based on the initial surface density of kerosene in the soil and the prevailing temperature. The C 9 –C 11 components of the kerosene had selectivities above zero and were thus removed preferentially, while the C 14 and C 15 components were characterised by negative selectivities. The selective volatilisation alters the composition of the kerosene that remains in the soil by increasing the concentration of the C 14 and C 15 components. The results for the different soil types indicate that the organic matter content of the soil affects the selective volatilisation, possibly through hydrophobic adsorption on surfaces, whereas the clay content appears to be less influential.

On the relation between viscosity and hydraulic conductivity for volatile organic liquid mixtures in soils

Abstract Changes in the volatile organic liquid mixture (VOLM) hydraulic conductivity in different soils are compared with corresponding changes in VOLM viscosity through an extended analysis of results from three previous experimental studies. The conductivity with regard to four different kerosene mixtures, corresponding to different degrees of volatilisation of the original kerosene, was determined in one set of soils; an increasing degree of volatilisation implies less lighter kerosene compounds, changing both kerosene viscosity and its chemical composition. In another set of soils, kerosene conductivity measurements were conducted at two temperatures, which provided two different viscosities but did not affect the kerosene chemical composition. Both volatilisation- and temperature-induced changes in kerosene viscosity and conductivity were studied in two of the soils. In all the soils that were used in the temperature experiments, the changes in kerosene conductivity could be successfully predicted by scaling the original kerosene conductivity value based on the observed viscosity ratio. For the chemically different kerosene mixtures, the changes in conductivity agreed with the corresponding viscosity changes only in inert sands. For a montmorillonitic loam, a montmorillonitic clay and a peat soil, considerable deviations were found between the conductivity ratio and the viscosity ratio; for the peat, which was also used in temperature experiments, no such deviations were observed at different temperatures. The deviations between the conductivity ratio and the viscosity ratio were also found to increase with increasing differences in kerosene chemical composition. These results indicate that chemical composition may be of major importance for VOLM hydraulic conductivity in interacting soils, apart from the effect that the composition has on viscosity. The viscosity ratios were shown to deviate more than 300% from observed conductivity ratios for the chemically most different kerosene mixtures.