Shale as a Permanent Barrier after Well Abandonment

Several thousand existing wells on the Norwegian Continental Shelf are to be plugged and abandoned in the coming years. It is a fundamental premise for the petroleum exploration activity that plugged and abandoned wells are properly sealed. This is however a time consuming operation, and the total expense is estimated to become somewhere between fifty and hundred billion US dollars. About 80% of this cost represents lost income for the State of Norway. A petroleum well consists, simply speaking, of a hole in the ground lined with a steel tube. In the most critical sections of the well, the annulus between the steel tube and the rock is filled with cement, in order to achieve proper sealing and to avoid leakage along the well. In some shale sections where the annulus is not filled with cement, sealing may occur as a result of the surrounding shale creeping in and closing the gap between rock and steel tube. If it occurs, this natural sealing process eliminates a significant portion of the work involved in creating a permanent barrier, and implies tremendous cost savings for plugging and abandonment operations. Moreover, the resulting shale barrier is probably more stable than a conventional cement barrier. Pressure tests show that shale may act as an efficient barrier in some but not all cases. As the mechanisms involved in this sealing process are not fully understood, there are currently no means for improving the likelihood of success. The main goal of this project is to acquire knowledge about the mechanisms involved when shale is forming a sealing barrier around a well. Furthermore, the project aims to establish procedures for predicting, and possibly improving, the efficiency of the method. The work involves a combination of theoretical analyses and laboratory testing, performed by SINTEF Petroleum Research with support from universities in Norway (NTNU) and USA. The project has performed a series of laboratory tests on shale cores, including cores from fields in the North Sea, confirming that shale barriers may be formed under given conditions. Whether this happens or not is controlled by the balance between the large compressive stresses that push the shale towards the well, and the pressure in the fluid in the well that pushes the shale outwards. In addition, most rocks have a certain ability to guide compressive stresses around cavities, which is favorable for building houses and tunnels, but unfavorable for the formation of shale barriers. This ability varies among different shale types, and consequently the probability for a shale barrier to form also varies with shale type. The ability to maintain a stable cavity depends on a combination of several rock properties. Moreover, a collapsing cavity is necessary but not sufficient for the formation of a sealing barrier. Prediction of shale barriers is therefore challenging. Verification of shale barriers in the field is also challenging, as standard leak-off tests do not have sufficient accuracy to reveal small, but significant leaks. It is possible to induce closure of the gap between rock and steel tube by varying pressure, temperature and fluid composition in the well. Results from laboratory tests as well as numerical simulations show however that as normal conditions are reestablished, as will happen after some time, the sealing effect is reduced even though the gap remains almost completely closed. On the other hand, rocks that are exposed to sufficiently large stresses may deform slowly over time (they creep). To the extent that this happens, it will contribute to a gradual improvement of the efficiency of the shale barrier. Moreover, some shale types have the ability to reestablish broken bonds, which implies that fractures may undergo a self-healing process, also improving the shale barrier efficiency over time.

- A general understanding of the main mechanisms for shale barrier formation. This is essential for efficient utilization of this technology and provide guidelines for completion procedures, and the ability to estimate the potential for shale barriers in specific locations. - A "shale barrier test" which allows for testing of shale barrier formation under field-like conditions on core material from the field. This test facility will play an important role in further development of this technology, including the search for reliable verification methods in the field. - A new model for creep in borehole geometry. This is valuable for the understanding and description of the shale barrier forming process. - Evaluation of pressure tests used to establish the integrity of an annulus barrier in the field revealed limited and possibly insufficient accuracy. This calls for standardization, and possibly development of new and improved test procedures.

A fundamental premise for all petroleum field operations is that the wells constitute the only flow channels between the reservoir and the surface. In the most critical sections of a well, the annulus between the casing and the rock is filled with cement in order to achieve proper sealing and avoid leakage outside the well. In some shale sections where the annulus is not filled with cement, sealing may occur as a result of shale creeping in and closing the gap between rock and casing. This natural sealing process eliminates a significant portion of the work involved in creating a permanent barrier, and it implies tremendous potential cost savings for plugging and abandonment operations. Pressure tests show however that this method works in some but not all cases. As the mechanisms involved are not fully understood either, there are no means or procedures that may improve the likelihood of success with this method. The main goal of this project is to establish methods for predicting, and possibly improving, the time dependent efficiency of shale as a self-sealing annular barrier around a well. This will be achieved by a combination of laboratory testing and modelling. The purpose of the laboratory tests is to identify how rock type and local, downhole conditions affect the efficiency of shale as a barrier. The modelling efforts will assist in the interpretation of the experimental results, tie links to shale properties that are normally available from field measurements like log data or cuttings, and build the methods for predicting and possibly improving the efficiency of shale as a barrier in given situations.