Reverse cementing: How can it work?

Primary cementing of casing strings is a necessary well construction operation where the annulus behind the string is displaced to a cement slurry. Cementing by reverse-circulation involves injecting cementing fluids directly to the annulus from surface, which can reduce the circulation pressure exerted onto the open-hole section during placement. We study density-unstable reverse-circulation displacement flows involving non-Newtonian fluids in near-vertical, eccentric annuli and use a semi-analytical displacement model to assess whether the interface between displaced and displacing fluids will elongate (unsteady displacement), or remain compact as cementing fluids are pumped down the well (steady displacement). We use the displacement model to study how casing eccentricity and fluid viscosity hierarchy impact the displacement classification. We find that typical field values tend to associate with unsteady displacements, particularly in regions of poor casing centralization, and for displacements involving "traditional" drilling mud, spacer and cement slurry properties. Improving casing centralization, increasing the displacing fluid viscosity and reducing the fluid density difference all act to stabilize the displacement. Similarly, increasing the imposed flow rate can also reduce the rate at which the interface elongates by enhancing the viscous stresses during placement. We compare selected scenarios to displacement simulations performed with the 2-dimensional gap-averaged (2DGA) model, and find qualitative agreement between the 2DGA results and our displacement classification predictions. While our study shows that it is possible to make density-unstable reverse-circulation displacements steady, we note that the identified measures are likely to increase the circulation pressures, which may offset an important benefit of reverse-circulation cementing.