Evaluation of Cement Systems for Oil and Gas Well Zonal Isolation in a Full-Scale Annular Geometry

Abstract

Abstract In a wellbore, loss of zonal isolation can be caused by the mechanical failure of the cement or by the generation of a microannulus. However, the behavior of the sealant is driven by the specific boundary conditions like the rock properties. Large-scale laboratory testing of the cement sheath in an annular geometry and in a confined situation was performed to simulate various well conditions and to evaluate the behavior of several sealants under simulated downhole stress conditions. The failure modes of the cement sheath were determined as a function of the cement mechanical properties, loading parameters, and boundary conditions. The results were used to validate an analytical model that predicts cement sheath failure. Introduction Interzonal communication in a wellbore may lead to loss of reserves, contamination of zones, production of unwanted fluids, or safety and environmental issues. Remedial solutions exist to repair the problems, but for technical or economical reasons, the well may be shut in or abandoned. To improve the lifetime of the well, the cement sheath must be chemically and mechanically durable. Sealants resistant to aggressive formation fluids are designed when required. In the same way, sealants should be designed to withstand the stresses experienced during production and well operations - e.g., casing pressure tests, stimulation treatments, or temperature changes during production cycles-throughout the well life. To achieve this, a better understanding of the mechanical behavior of different sealants under downhole conditions is required to design fit-for-purpose materials.1,2 Several papers have been written on the subject. According to Thiercelin et al.,3 changes in downhole conditions can cause mechanical damage to the cemented annulus (mechanical failure or creation of microannuli) that may lead to a loss of zonal isolation. The key conclusion of that paper was that instead of considering the strength of the sealant as the main property, one should rather look at the complete mechanical system formed by the steel casing, the cemented annulus, and the formation. Indeed, increase of pressure and/or temperature in the wellbore firstly expands the inner steel casing, which instantly imposes this deformation on the neighboring cement sheath. As a consequence, imposed displacements rather than imposed stresses are applied to the cement inner diameter (ID). At a greater time scale (the lifetime of the well), the cement sheath must withstand multiple displacement cycles. Several authors have proposed numerical models4,5 to simulate the sealant mechanical behavior and predict initiation of failures according to known mechanical properties of the complete system (steel, cement, and rock). A large-scale laboratory test for sealants in an annular geometry has been developed. Changes in the well conditions resulting in either the contraction or the expansion of the inner casing can be simulated. Furthermore, the confining role of the formation or outer casing can be evaluated. Such an experiment allows the evaluation of the sealant mechanical response under wellbore conditions. Indeed, the nature of stresses generated in the annulus (tensile and/or compressive) is similar to those the sealant must withstand in a real wellbore. The loading scenario simulated in the full-scale annular sealing test is close to reality. Several cement systems exhibiting different mechanical behaviors have been tested, and the experimental results have been compared with the predictions of a numerical model. Laboratory experimentation The experiments are designed to compare different cement formulations at room conditions in a large-scale annular geometry and determine the effect of cement mechanical properties and boundary conditions (rock stiffness) on cement cracking and permeability to air. Imposed deformations can be applied on the cement ID to simulate changes in wellbore conditions caused by variations of temperature and/or pressure. Equipment The equipment developed for the study is shown in Figs. 1 and 2. There are two main components.