Experimental studies about solute transport in soil are generally conducted under saturated conditions, whereas studies with unsaturated media are usually limited to hydrodynamic analysis. Only a few studies focused on the link between unsaturated flow and physical, chemical and biological interactions, which are controlling pollutant availability. However, the presence of a gaseous phase in soil can modify not only the movement of soil solution, but also chemical and biochemical interactions and exchanges between soil aggregates and solution. Study of reactive solute transport in the vadose zone appears then to be a necessary stage to predict contaminant fate in natural soils, for risk assessment as well as for the design of effective remediation processes for contaminated soils. This question is the main objective of the present work developed in the frame of the French Scientific Interest Group Industrial Wastelands called “GISFI” (www.gisfi.prd.fr), based around a scientific and technological project dedicated to acquisition of knowledge for sustainable requalification of degraded sites polluted by past industrial activities. The aim of this work is the study of contaminant transport in variably saturated soils at the bench scale. We will compare reactive solute transport in saturated and unsaturated flow to evaluate the pertinence of the studies conducted in both conditions. This research is mainly methodological, thus model systems are chosen at first in order to study ion exchange and Polycyclic Aromatic Hydrocarbons (PAH) migration. These pollutants are indeed among the most widely discussed environmental contaminants because of their toxicity for human health and ecosystems and they are present in large quantities in soils polluted by former industrial activities.

Experiments are performed at the lab scale with soil columns allowing a good control of operating conditions and following the methods of frontal chromatography. Saturated flow experiments are conducted in a conventional bench-scale column apparatus, including a pump, a glass column, on-line detectors (pH and conductivity) and data acquisition. A specific soil column system is used to carry out experiments with an unsaturated steady-state flow. This set-up is composed of two stainless steel columns (outside diameter: 10 cm; length: 20 cm), equiped with two tensiometers, with a sprinkler unit and a stainless steel porous plate at the bottom end. Suction can be applied thanks to a vacuum pump and a multichannel peristaltic pump allows the feeding of the columns and the depletion of the outflow liquid. Experiments are monitored thanks to a specific software. In both cases, samples are collected at the column outlet for further analysis (PAH, ion analysis, TOC).

Different porous media were selected: a sand, a sand coated with kaolinite, a sandy-silty soil sample, a contaminated soil sample (PAH, heavy metals) from a former coking plant. For each sample, the soil-moisture characteristic curve (representing matrix suction versus water content) is experimentally determined in order to find the best operating conditions for our experiments with unsaturated flow. Numerical codes allowing us to model reactive solute transport under variably saturated water flow conditions are also used to model the results.

The results are currently in acquisition and will be presented.

This study at a bench scale aims at developing at first a methodology under unsaturated conditions and better understanding the main mechanisms controlling reactive solute transport in natural soils. This is a necessary preliminary step before upscaling at the lysimeter and field scales. Modelling of the observed processes will enable us to predict long term fate of contaminant like PAH in soils.

Keywords: solute transport, column experiments, vadose zone, bioavailability, organic contaminants