Deep magmatic degassing produces metalliferous supercriticals fluids and brines that accumulate above magmatic reservoirs at depths of 1 to 6 km beneath active volcanoes. These brines can be exploited for the metals they contain and for geothermal energy. To quantify the volume of brine present in active volcanic centers from magnetolluric data, we need to establish conductivity models of brines by varying pressure, temperature and composition. To begin with, we are working on completing the conductivity database for the H2O-NaCl system under PT brine conditions and developing a theoretical model to explain these data. The current database is based on work by Marshall et al.(1968), Sinmyo et al.(2017), Guo et al. (2019) in P-T range 0.1MPa-5GPa and 0-900C. We planned to do experiments to cover the range 20MPa-120 MPa and 0-900C. To do this we will fill a porous mullite with brine and measure its conductivity by Electrochemical Impedance Spectroscopy under PT conditions in the Paterson press. The brine conductivity depends on the number of ions Na+,Cl-,H+,OH- and the diffusion coefficient of each ion, these parameters being dependent on pressure, temperature, pH and initial concentration of NaCl. To calculate speciation we use the Deep Earth Water model developped by Sverjensky et al. (2019) based of Helgeson-Kirkham-Flowers equations of state. Ion diffusion coefficients are calculated using Debye-Hückel-Onsager theory and are dependant of viscosity and dielectric constant of the solvent, temperature and ionic strength.