Setting Rheological Targets for Chemical Solutions in Mud Removal and Cement Slurry Design

Abstract

Abstract A major application area for chemistry-based solutions in oilfield cementing is in the reliable design of fluid rheologies for different purposes. A common situation is that one viscoplastic fluid will be displaced by another fluid along some sort of channel, e.g. a pore-space or the annulus during cementing. Whether the resident fluid is completely displaced will depend largely upon the fluid rheologies and hence on the chemical design of the fluids. Understanding the mechanics of this process enables one to set rheological targets for applied chemistry solutions. We present new results of university and industrial research into the displacement of visco-plastic fluids along long ducts, (pipes and slots). The research combines laboratory experiments with computational studies and with detailed mathematical analysis of the fluid mechanics. The results are both new and surprising, being in many respects counter to accepted oilfield intuition. The direct application of these results is in the design of spacer and cement slurry properties for effective mud removal during primary cementing. The breakdown of zonal isolation can sometimes be attributed to poor bonding of the cement to the casing and formation, due to the existence of residual mud layers on the walls of the annulus after cementing, i.e. the formation of a so-called "wet micro-annulus". It is possible to detect such layers using advanced ultrasonic cement evaluation logs. A visco-plastic fluid displaced by a fluid with a smaller yield point can leave a static residual layer of gelled fluid on the walls of the channel. We show that it is possible to predict the maximum static layer that can remain. Actual static layers that are observed in simulations and experiments are much thinner than the maximum layer. Additionally, the variation in static layer thickness with rheological and process parameters is not at all intuitive. For example, increasing the mean velocity can actually increase the static layer thickness and increasing the yield stress of the fluid to be displaced can result in a thinner static layer! These results are confirmed by laboratory experiments, computational simulation and mathematical analysis. The results are quite novel. Displacements with viscous fluids do not leave fully static residual wall layers, unless other physico-chemical phenomena are present. Similarly, previous oilfield investigators have considered that, in order to remove a layer, it is necessary to design the fluid rheology so that the maximum static layer predicted has zero thickness. We argue that this is rarely necessary, although sufficient, and that static wall layers can be removed with reduced rheologies. The most significant aspect of this research is that finally, with knowledge of local displacement velocities and fluid properties, we are able to predict the likely risk of a wet micro-annulus occurring during a primary cementing operation. Conversely, we are able to make recommendations for rheological fluid properties that are sufficient to avoid a wet micro-annulus and thus to enhance the prospect of complete zonal isolation during primary cementing. This sets the targets for applied chemistry solutions.