Physical Effects Controlling Mass Transfer in Matrix Fracture System during CO2 Injection Into Chalk Fractured Reservoirs
MetadataVis full innførsel
Tertiary recovery or Improved Oil Recovery (IOR) methods are key processes to replace or upgrade reserves, which can be economically recovered, beyond conventional methods. Therefore, the application of IOR methods offers opportunities to increase the hydrocarbon reserves that have been produced in addition to those coming from exploration and reservoir appraisal. The purpose of this thesis is to combine experiments, computations, and theory to make fundamental advances in our ability to predict transport phenomena as well as the IOR potential involved in tertiary CO2 injection at the lab scale in a matrix fracture system. This is done by using rock and fluid samples similar to one of the chalk fractured reservoirs in the North Sea. The work involves a review of key physical mechanisms and calculation methods for the modelling of fluid flow in fractured reservoirs. The main matrix fracture fluid exchange mechanisms described are gravity drainage, capillary imbibition and molecular diffusion. Also described are the estimation of the recovery performance for a single block and a stack of blocks surrounded by gas. The effect of interfacial tension on the ultimate recovery has been discussed and the definition of the minimum miscibility pressure for single porosity and dual porosity system is described. Numerical modelling of gravity drainage for a matrix blocks surrounded by gas has been described. Numerical estimation of gas-oil gravity drainage by reducing the number of grid blocks in vertical direction in a draining matrix column is common practice in order to reduce the simulation time. However this can lead to systematic numerical errors and consequently underestimation of the recovery. In order to minimize the underestimation of the reservoir performance, a set of pseudo functions needs to be developed that not only satisfy the actual responses in the fine grid simulation but also reduce the simulation time. The effectiveness and the accuracy of such pseudo functions are extensively discussed and the different simulation models have been run to quantify the underestimation of recovery by coarse griding in the numerical modelling of gravity drainage. The importance of the molecular diffusion to recover oil from a high fracture intensity system is described as well as the basic concept for calculating the molecular diffusion based on the Fick’s second law. Corresponding laboratory methods for the estimation and measurement of the oil and gas diffusion coefficients are reviewed. The effect of molecular diffusion on the interfacial tension and eventually on the gas-oil capillary pressure is presented. A compositional study of a non-equilibrium gas injection process such as CO2 requires an equation of state (EOS) model which can predict the CO2/oil phase behaviour. In order to make such EOS model, a set of pVT experiments using fluids involved in the core flooding has been performed and finally the EOS models were tuned against experimental pVT data. The necessary steps to perform pVT experiments including making live reservoir oil, constant composition expansion, single flash, viscosity measurements and CO2-oil swelling are described. Gas injection is known to have a significant potential for high ultimate recovery in many oil fractured reservoirs with tall matrix blocks. The high ultimate recovery in these reservoirs could be due to the effectiveness of the gravity drainage mechanism. Fractured chalk reservoirs in the North Sea have a very high porosity (up to 45%), and low matrix permeability (3-4 mD) with small matrix block size. In order to quantify the dominant transport mechanisms and potential of Improved Oil Recovery (IOR) in the case of CO2 injection in the North Sea chalk fractured reservoirs, CO2 injection experiments at reservoir conditions have to be performed in the laboratory. The feasibility of such laboratory experiments initially has been verified by performing compositional simulation. In these simulations by varying the experimental parameters, such as core height and fracture size, the optimum matrix and fracture geometry were designed and the summary of the task is presented in Paper 1- Appendix A. CO2 injection experiments under reservoir conditions in the presence of different water saturation at reservoir conditions have been carried out. A unique technique has been developed for saturating the matrix system with reservoir fluids. This method ensures a homogeneous fluid composition within the pore system before the fracture system is initialized with the CO2. A complete description of, rock and fluids, experimental procedure and experimental results is given in Chapters 3, 4 and Papers 2 and 3 in Appendices B, C. In order to investigate the effect of temperature on the oil recovery mechanism, CO2 injection experiments were carried out at initial reservoir temperature (130 ºC) and a low temperature 60 ºC which representing the water flooded parts in the reservoir. The effect of initial water saturation also was investigated at reservoir temperature 130 ºC by performing two experiments with different initial water saturation. Results from these experiments show a high potential for oil recovery in all experiments. In the high temperature experiments, the produced oil had a variable composition during CO2 injection, while at the low temperature condition, the produced oil initially had a constant composition and then it started to change. Different behaviour of produced oil composition in the high and low temperature might be due to dominant of diffusion mechanism in the high temperature experiments. In the low temperature (60 ºC) experiment, at the early stage of CO2 injection the produced oil had constant composition for a short period of time and then it changed to variable composition similar to the high temperature case. This behaviour maybe is due to high solubility of CO2 into oil and consequently more oil swelling than the high temperature condition. In order to quantify the above mechanisms, several attempts have been done to history match the experiments by using compositional simulator. But in all cases, it was not possible to history match the experiments. The weakness of the simulator was due to the improper formulation which was used for calculating the cross phase diffusion between the oil and gas phase in the matrix and fracture system. The details of simulation work as well as the cross phase diffusion issue are discussed in Chapter 5 and Paper 2 in Appendix B.