Reservoir Conditions Laboratory Experiments of CO2 Injection Into Fractured Cores

Abstract

Abstract The efficiency of tertiary CO2 injection at the reservoir conditions into fractured cores has been investigated experimentally. The experiment was designed to illustrate the process of water imbibition and CO2 injection into a North Sea chalk reservoir. The core and core holder assembly were designed to allow a 2 mm gap to surround the core simulating a fracture. Live reservoir oil was prepared and used to saturate the matrix system. Due to the large permeability contrast between the core (4 mD) and the fracture it is not feasible to saturate the core by simply flooding the system with live oil, since oil would flow through the fracture and only partially saturate the core. To overcome this problem 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 by the injection fluids (water/CO2). During the experiment, the matrix system was first saturated with the live reservoir oil, during which the gap was blocked by a sealing material. In the next step the sealing material was removed and water was injected into the gap. Finally CO2 was injected at a low and constant rate into the gap. Oil and water production and fluid composition were monitored and the results show that injection of CO2 could significantly recover residual oil after water injection. Introduction To generate accurate predictions by field simulations of CO2 injection into a fractured reservoir, a set of CO2 injection experiments at reservoir conditions should be performed to check the compositional effects on the displacement process and to quantify the most important mechanisms. The results obtained can then be up-scaled to larger block sizes and finally allow development of the necessary transfer functions for any field scale simulation. Performing laboratory experiments at reservoir conditions on fracture systems faces many challenges. The most important is to saturate the matrix and fracture with representative reservoir fluids. Therefore in previous experiments cores were not saturated with live reservoir oil. For example in the CO2 gravity drainage experiments performed by Li et al.1, the core was saturated with dead oil while the experiment was supposed to be at the reservoir conditions. Dry gas injection in fractured chalk by Øyno et al2 was conducted by saturating the matrix system with live oil, but still their method for initialization of the pore system with live oil is not certain. In their experiment the oil recombination was carried out in the core holder where the matrix and fracture were placed. The oil/gas mixture was circulated in the system and pressure was monitored. Once pressure had stabilized, they assumed that the pore system is saturated with the live oil. In this method, since the pore system was saturated with oil by a very slow diffusion mechanism, pressure stabilization over a short time interval will not guarantee homogeneous initialization of the pore system with representative reservoir fluids. The main objective of this work has been to investigate the efficency of the tertiary CO2 injection into a fractured core, focusing on the mass transfer of CO2 and hydrocarbons between the fracture and the chalk matrix. The resulting recovery profiles and produced fluid compositions can later be used to construct a compositional numerical model. Using this model the magnitude of all forces as well as their contribution to the displacement mechanism during the experiments can be studied. An important feature of CO2 injection into fractured reservoirs is the potentially large component exchange between fracture and matrix system. This requires an accurate equation of state (EOS) for modelling purposes. Therefore a comprehensive fluid study of the CO2/oil system including swelling tests has been performed. Experimental For a reservoir with a long water injection history, the reservoir temperature will typically be reduced. The highest reductions are seen in the proximity of the injection wells. In the field addressed by this study the initial reservoir temperature is reduced from its initial value of 130 ºC. For this experiment 60 ºC was chosen to represent the temperature of the water flooded zones. The current reservoir pressure is 300 bar. The full detail of the core samples, fluids and experimental set up are described below.