In the last years, the detection and monitoring of contaminants in non-aqueous phase commonly called non-acqueous phase liquids (NAPLs) are becoming very important for the remediation of contaminated sites. The NAPLs are divided into two main categories: the light-NAPLs (LNAPLs) which include light pollutants with a density less than water and dense-NAPLs (DNAPLs) with a density greater than that of water. LNAPs are part of the aromatic hydrocarbons such as toluene, benzene and derivatives, xylene, etc., which tend to form "pools" and spread laterally in the presence of water because of their low density (Lesage and Jackson, 1992), in fact, in case of contact with an aquifer they tend to remain insoluble on the surface of the aquifer becoming a danger of contamination to the environment. DNAPLs are part of the chlorinated solvents, compounds derived from hydrocarbons by aliphatic and cyclic hydrocarbons such as tetrachlorethylene (PCE), trichlorethylene (TCE), carbon tetrachloride, etc.. They are for the most part, a very good solvent power, propellant, refrigerant and low flammability. Due to their characteristics, are widely used in chemical, textile, rubber, plastics, fire extinguishers, coolants, in degreasing and cleaning of metals, leathers and fabrics. The DNAPLs can reach considerable depth because of their higher density than water and are considered the most common cause of contamination of the subsoil. After their release into the environment, the DNAPL migrates, by gravity, through the vadose zone and, soon as it encounters in its path the aquifer, because of its high density, tends to move downwards, until it finds one low permeability layer. In its downward movement, part of the DNAPL remains trapped between the pores of the medium passing through, creating bodies of discontinuous (residues). When passing through the flap, part of the DNAPL, still in phase separate, dissolves contaminating the aquifer. The DNAPL, it is usually not consist of a single chemical component, for which, in water, the various components will be dissolved in various quantities according to their own solubility. The dissolved phase of the DNAPL, therefore, moves with the movement of the groundwater flow going to contaminate in this way also areas very distant from the initial point of release. The hydro-geological characteristics of the site, joined to the unstable behavior of the DNAPL, causes the condition of a site contaminated by DNAPL is complex comprising pollutants being separated, dissolved and gaseous or to the exclusion of the gas phase can contaminate the subsoil for long periods (Illangasekare et al, 1995). Although the process of mass transfer at the water / DNAPL is well known, the process that occurs in natural systems under real complex morphologies and hydrogeological situations, continues to be the subject of many studies (Page et al., 2007). The distribution of the DNAPL is typically influenced and controlled by the heterogeneity of the porous medium involved (Shwille, 1988; Kueper and Frind, 1991), then in order to understand in detail the behavior of the DNAPL inside the medium, there is a need detailed knowledge of the subsurface. There have been many studies regarding both laboratory scale (Power et al., 1998; Saba & Illangasekare, 2000) and at the level of theoretical models (Nambi & Powers, 2003; Bradford et al., 2003; Parker & Park, 2004). In any case, the influence of the heterogeneity of the medium on the migration of the DNAPL still remains uncertain (Soga et al., 2004; Fure et al., 2006). Even in theory, the interpretation of the phenomenon is complex because it is impossible to recreate the real distribution of DNAPL in the subsurface (Brusseau et al., 2007). In fact, in the soil, the separate phase of the DNAPL may not be present as a continuous phase and the extension of the contaminated area, the separate phase and the dissolved phase, is highly dependent on physical-chemical properties of the medium and of the contaminant. Both phases, separated and dissolved, may be present in the same area in different percentages and the transition zone, between the contaminated area and the non-contaminated, depends only on the physical characteristics of the soil and of the fluids contained in it. The source of contamination, the extension of the plume of contamination and saturation of the DNAPL in relation to the depth are usually determined using monitoring wells, coring, etc.. In addition to these direct methods, however, are attributed to high costs and considerable disadvantages due to the worsening of the situation of contamination of the site, as they may convey the DNAPL from deep to superficial layers, provide timely hardly be extended to large areas, etc.. It is clear that a detailed characterization and precise monitoring of the sources of contamination are essential points for a reclamation project (Kavanaugh et al., 2003). And for this, where there is significant heterogeneity of the subsoil, the conventional direct methods and intrusive sampling in groundwater may be insufficient, since the information they gain is restricted to vertical profiles and sampling point and not extended to entire area contaminated (Chambers et al., 2010). To try to solve these problems, in the last years are increasingly claiming the use of indirect methods for the detection and monitoring of sites contaminated by DNAPL (U.S. EPA, 2004). These methods have the advantage of producing information on the entire contaminated area at low cost and in a relatively short time being sensitive to changes of the physical properties of the fluids which go to investigate. To date, the research on the applicability of these methods, has proved their worth in the detection of DNAPLs, but has not yet been able to verify which of the different phases of the DNAPL (separate, dissolved and gaseous) you can actually receive and distinguish if actually may be able to give answers selective on the phases of the contaminant. One of the objectives of this thesis is precisely to verify, through the integration of different electromagnetic methods (Ground Penetrating radar, electric measurements and permittivity measurements (Time Domain Reflectometry)), the effective potential of some of the electromagnetic methods in the detection of the different phases in which DNAPL divides the inside of a porous medium saturated. For this objective, two experiments were carried out in the laboratory, in a controlled environment, using two different devices. For both experiments, it was decided to use a homogeneous porous medium saturated, placed inside the tanks. DNAPL was used as a chemical is not toxic (HFE-7100) as similar physical characteristics to trichlorethylene (TCE). In the first experiment were carried out the tests only with the GPR method; in the second experiment, instead, were carried out all the tests listed above: GPR, electrical measurements of resistivity and direct measurements of permittivity through the TDR method. For what concerns the GPR it was decided to use an antenna multicomponent 4 channels placing another objective which is to check if there might be a preference in the detection of the DNAPL through a channel rather than with another, evaluating the extent to which the components of the electromagnetic field could be sensitive to the presence of the contaminant. In Chapter 1 of this thesis highlight the main aspects related to the state of the art on the study of contaminated sites, it examines the complex dynamics of migration of DNAPL in a porous medium in terms of physico-chemical and shows the importance of the study of theoretical models as a complement to assist the interpretation of real data. In Chapter 2 there is a hint on the fundamentals of electromagnetic theory of the different methods used in the two experiments referring insights in basic texts In Chapter 3 we explain in detail all the materials, methods, instrumentations used in two laboratory experiments. I describe the preliminary laboratory tests that were conducted prior to the performance of the experiments and show the results of theoretical models developed for both measures that GPR for electric ones. The theoretical model GPR is primarily used to study a multiphase system (air, water, sand, DNAPL), quite similar to the real case, creating cases at different saturation conditions, to help interpret and validate the results obtained from the real case. With the theoretical model electric, instead, it is attempted to understand what might be the edge effects induced by the small size of the devices used. Through the model, in fact, it was decided to analyze the data relating to the minimum distance electrode in order to avoid edge effects linked to the walls of the device. It is also seen as a margin of error there might be to consider the apparent resistivity measured as real resistivity. In Chapter 4, have been presented in detail all the results in the two experiments. As regards the results GPR were analyzed the variations of the delay times of the reflections of the electromagnetic wave inside the tank and the spectra of the amplitudes for the 4 channels of the antenna. For electrical measurements, have been mapped variations in apparent resistivity inside the tank in different moments of the experiment. The TDR measures, however, have served to have a direct measure of changes in permittivity (ε) in the neighborhood of the point of entry of contaminant during the various phases of the experiment. In Chapter 5, finally, we have drawn the conclusions of the work of thesis highlighting how, through the use of different electromagnetic methods, we can begin to have a clearer landscape on the potential of these geophysical methods (GPR, and electrical measurements TDR) for the interpretation of the complex scenario that is created in a site contaminated by DNAPL. The simultaneous use in a controlled experiment of these methods electromagnetic led to understand that, through a thorough study of the variations in physical properties of the contaminated medium we can distinguish the dynamics of the slow migration and transformation of the contaminant within the vehicle and can be perhaps arrive to discriminate the different phases of the contaminated (separate, dissolved and gaseous). This is certainly a first step towards a better understanding of the complex migration behavior of DNAPLs can be considered as an advance on the knowledge of the responses of these methods and a starting point for future developments.