Wellbore integrity is a concern especially in regard to leakage potential of the annular cement sheath. The annular cement sheath is considered a primary barrier of fluid migration; however, it is often only partially cemented in active wellbores. Regulations for the cement height, known as top of cement (TOC), vary from region to region and for different applications yet the potential of long-term wellbore integrity is often not considered in regulations. Finite element models depicting the well's life cycle have shown that the annular cement sheath experiences tensile debonding at the casing interface when the wellbore experiences a pressure reduction which occurs after a reservoir is depleted. The microannuli that occur from cement sheath debonding have recently been quantified, but for thin sections at discrete depths. This paper quantifies the microannuli that can occur throughout the casing string up to the TOC line while considering the various wellbore parameters through the use of a finite element model that includes gas hydraulic microannuli fracture propagation. A case study of a Gulf of Mexico was performed and showed that the cement sheath can experience debonding up to the TOC line when gas is the leaking fluid whereas when water was simulated, the fracture did not propagate to the TOC line.
For all the potential leakage pathways in a wellbore, the cement used when completing and abandoning wellbores is the most at risk for integrity issues. The steel casing has been well studied and is designed with safety factors and design considerations for extreme conditions in the Gulf of Mexico. The surrounding rock formation has also been studied extensively however, there are less design modification engineers can do for long term integrity with respect to the formation. The cement used when completing and eventually plugging the well is less understood with many design elements than can change its properties. Numerical models that predict cement sheath integrity assume the leaking fluid is incompressible and have omitted gases from their results (Zielonka et al. 2014; Feng et al. 2017; Kumar et al. 2017; Searles et al. 2018; Bois et al. 2019; Jiang et al. 2020; Xu et al. 2020; Gheibi et al. 2021). The methods that do investigate gases, have assumptions that are unrealistic for wellbore conditions. Al Ramadan et al. (2019) accounted for an ideal gas while Corina et al. (2020) analyzed gas flow data using Hagen-Poiseuille flow. Other assumptions, within models described by Ford et al. (2017), Lavrov and Torsaeter (2018), Moeinika et al. (2018), and Al Ramadan et al. (2019), are that microannuli are constant arbitrary values along the length of the cement.
