Abstract
It has been demonstrated that creeping shales can form effective hydraulic well barriers. Shale barriers have been used for many years in P&A of wells in Norway. More recently, shale barriers for zonal isolation have also been used in new wells where shale creep was found to occur within days. In some cases, shale creep is activated by a reduction in annulus pressure, in other cases shale creep sets in without any active activation, possibly by time-dependent formation-pressure changes. However, the presence of thixotropic fluids (drilling muds) in the annulus may prevent full closure of the annulus as it requires large pressure differentials to squeeze the fluid out of a microannulus. Furthermore, elastic rebound of an actively activated shale barrier could result in a microannulus and hence a possible leakage pathway. Improved logging technology is needed for identifying shale barriers and the presence of micro-annuli in shale-barrier zones.
We use cement bond log data and standard bond logging criteria to evaluate the quality of the shale well barriers (Williams et al., 2009). In addition, in order to detect microannuli on the outside of the casing, a new inversion algorithm for the bond logging data was developed and tested on field data. Later, we had the chance to apply the inversion algorithm to bond-log data obtained in the laboratory with a miniature bond-logging tool inside a cased hollow-cylinder shale-core sample place. It turned out that both the micro-annulus widths and shale velocities determined by the inversion technique were too high. By constraining the shale velocities to more realistic values, the updated microannulus widths were smaller and more consistent with the experimental results.
Small microannuli may not cause any measurable leakage along the well, especially if filled with a thixotropic fluid. However, more studies are needed to quantify the impact of microannuli on the sealing capacity of shale barriers.
