No abstract

When gravel packing wells in brine after an oil-based drill-in, filter cake strength and permeability are important factors for successful operations. These relate to controlling fluid losses, firstly, by resisting mechanical wear when running production screens and, secondly, by providing a tight hydraulic seal when pumping gravel.Two oil-based mud formulations with "initial" and "depleted" bridging packages, respectively, and otherwise equal formulation, were used to deposit filter cakes on outcrop sandstone plugs in a Hasslertype core holder. Drilling mud was then displaced by gravel pack carrier brine keeping an overbalance pressure. Brine was allowed to permeate through the oil-based filter cake. The filter cakes before and after exposure to brine were compared in terms of cake strength, and volume of brine produced through.It was found that brine transport through the filter cakes at the same overbalance pressure was slower than the static mud filtration rate. Scratch tests showed a higher strength for the filter cake formed with the "initial" particle size distribution, and that both filter cakes strengthened significantly following brine diffusion. NMR T 2 distributions and 1D NMR intensity profiles showed that core plugs exposed to the mud formulation with "depleted" bridging package experienced stronger alterations in NMR properties and deeper damage than the formulation with the "initial" bridging package.The methods used in this study have shown to have the potential of probing some of the mechanical and hydraulic properties of filter cakes and the extent to which core plugs are invaded by mud particles during filtration and filter cake buildup. Future work will address the effect of different base fluids, water-based or oil-based, and of different bridging packages to improve our understanding of how the fluid basis affects filter cake qualities. Other options are to look at the effect of wash pills on filter cake properties.

Sand onset and sand rate predictions are important in hydrocarbon production to optimize production, increase recovery, and reduce costs and the environmental footprint. Recent laboratory results on Castlegate sandstone from sand production tests in a True Triaxial test system have revealed that stress anisotropy plays an important role not only on sand onset but also in sand rate. The results confirmed our hypothesis that stress anisotropy means earlier sand produced but less sand. The laboratory results also revealed the effect of fluid saturation, i.e., oil, brine or irreducible water saturation on sand onset and sand rate. They allow the calibration of SandPredictor, a field sand prediction model, for stress anisotropy and production before and after water breakthrough. A field case analysis demonstrated the effects and showed the importance of in situ stress anisotropy and watercut on sand mass and rate.

Scratch testing to determine rock shear strength has been adapted to soft materials such as filter cakes. A soft metal blade with a strain gauge attached to it and calibrated to obtain the deformation to force correspondence replaces the conventional cutter in the scratch apparatus. The shear strength is obtained from the tangential force reading when the blade is moved at a constant applied velocity across a filter cake, for a given cutting depth. In this study, four commercially available drilling mud types (2 water-based and 2 oil-based drilling fluids) were filtered on ceramic discs. The resulting filter cakes were then scratch tested at various imposed cutter blade velocities, to determine the cake shear strength. The goal of this study was to investigate whether the scratch apparatus gives different strength values as a function of the cutter velocity. The results from this measurement campaign, where also a sandstone was tested, show that for the range of available cutter velocities, no effect can be detected on the computed strength values. The results thus suggest that no rheological effects such as shear-thinning occur at these velocities (for the filter cakes), as well as no undrained effects on the pore pressure. This is good news as it validates all previous scratch campaigns, removing the need for reviewing the relevance and interpretation of the scratch test, especially as applied to filter cakes.

No abstract