Since industrial revolution, due to the increasing demand of energy, anthropogenic emissions in the atmosphere are constantly growing. The International Energy Agency (IEA) predicted a 57% increase of energy demand from 2004 to 2030 (IEA, 2004) of which, 85% consists of fossil fuels. Actions need to be taken in order to mitigate CO2 emissions generated through human activities. In the last twenty years CO2 emissions became a political and industrial priority in many governmental and commercial institutions; Carbon Capture and Storage (CCS) is one of the main options but it is a temporary solution. CCS is based on capturing CO2 from large point sources. The gas is transported via pipelines and injected in deep underground formations, such as depleted gas and oil fields, deep saline aquifers and unminable coal seams. The last method is the option that will be treated in this thesis. Coal can be favorable if CO2 replaces coal gas that mostly consists of methane (CH4). This is called CO2- Enhanced Coal Bed Methane (CO2-ECBM) production. The feasibility and economical viability of CO2-ECBM depend on geological factors such as heterogeneities of coals and their pore systems. This is why the development and implementation of reservoir simulators for ECBM production and CO2 storage require detailed and reliable information on the physical and chemical processes that are initiated by injection of gases in the coal layers. The most important processes to deal with are sorption behavior, of the coal competitive sorption of the different gases presents, multi-phase transport, permeability behavior and initial amount and compositions of the gas injected. In this study the main objective is to get a better understanding of the coalwater- gases system. The activities involved experimental work and theory development to acquire data and theory for field scale modeling. Thinking in terms of real case scenarios we consider the use of an impure CO2 stream, i.e., impurities in the CO2, either flue gas components or water, and their effect on coal behavior. The activities in this research involve the experimental results of sorption on dry and wet coal and the sorption of flue gas type of gas mixtures measured with a manometric set up at 318 K and up to 160 bar. A new aspect of the thesis is to focus on the different ways to obtain sufficiently accurate Equations of State (EoS) to be used in a manometric set up. The experimental results allowed us to interpret and test different models concerning sorption and thermodynamic behavior of gases. The results in general show that sorption and desorption of CH4 and N2 on coal are fully reversible, meanwhile this is not happening for CO2. The equilibration time for the CO2 sorption on coal is much larger than for N2 and CH4. An increase in temperature is negatively affecting the sorption capacity of the coal. The swelling induced by CO2 injection on coal is a fully reversible phenomenon and it is positively related to the sorption. The sorption of CO2 on wet coal is inhibited by the presence of water. The density of the CO2-H2O gas phase in the temperature and pressure range of the study can be calculated using the Span and Wagner EoS for pure CO2. The CO2 dissolved in water, assuming that water in its sorbed phase behaves as in its free phase, can be described by a Peng-Robinson-Stryjek-Vera EoS that is optimized for the CO2-water system. The adequacy of an EoS to predict the density of a mixture can be tested by using a combination of the manometric set up with a density meter. A Helium mixture containing 1% O2 and 1% NO2 can be described with the Mc Carty EoS for pure Helium. In this case, the maximum relative difference from the experimentally determined density is of 6·10?3. A CO2 mixture containing 1% O2, 1% He and 1% NO2 cannot be described accurately with any of the existing EoS. Results concerning the excess sorption isotherm are influenced by the choice of a specific EoS. The maximum of the excess sorption can vary 25.88%, depending on which EoS is used for the calculations. A combination of different EoS for different pressure ranges gives an accurate result, with a maximum relative difference from the experimentally determined density of 0.05. In this case the maximum excess sorption gives a value of 7.81 mol/kg, which is in agreement with the literature concerning pure CO2 sorption on activated carbon. In the desorption process from activated carbon with the two mixtures mentioned previously, the mass spectrometer measurements show that for the specified P,T range, no reactions occur between the gas and the activated carbon. The results of this research give an alternative direction with respect to the use of impure CO2 in ECBM.