Mechanical Testing and Modelling of Oil & Gas Well Cement Sheaths

Hydrocarbons, such as oil and gas, are often found in underground reservoirs. Boreholes are drilled into the reservoirs to extract these resources and steel pipes, named casings, are placed inside the holes to pump the oil or gas to the surface while maintaining the stability of the hole. The gap between the borehole and the casings is filled with cement, creating the cement sheath. The cement sheath is crucial for well safety and environmental protection as it prevents uncontrolled oil and gas leaks outside the production pipes. However, the brittle nature of cement can lead to the development of cracks, leading to loss of well integrity and causing leaks that potentially pollute underground water reservoirs and the surrounding environment. This thesis focuses on analysing the mechanisms and consequences of cracking development in cement sheaths that lead to loss of well integrity.

The existing literature reports crack widths as wide as 500 µm, but there is a lack of experimental results to quantify the presence and relevance of cracks below 200μm. Therefore, a methodology based on Digital Image Correlation (DIC) and a novel data analysis process to automatically detect and quantify cracking in laboratoryscale was developed during the PhD project, targeting crack widths in the range of 5 to 200 μm. The methodology also allowed the evaluation of crack characteristics such as position, length, and orientation, which are subsequently used to calculate parameters such as the spacing between cracks and the magnitude of cracked areas. The developed methodology is employed to document and quantify crack development in cement sheaths under two specific loading conditions: i) cracking due to restrained shrinkage of the cement, which induces tensile cracks, and ii) cracking due to the steel casing expansion, which induces tensile cracks and plastic deformations due to high compression loads.

Regarding restrained shrinkage-induced cracking, which is one of the most significant causes of crack development in cement sheaths, an experimental investigation was conducted to quantify how extensive the damage might be due to changes in ambient humidity conditions, confinement levels, and mixture compositions. The presence of lateral confinement due to the presence of shale provided an 80% reduction in the cracked area compared to unconfined specimens, but still radial cracks as wide as 200 µm, microannuli up to 100 µm, and cracked areas of 50 mm2 per representative well cross-section were observed. Additionally, a reduction in crack widths and potential leak paths was achieved by reinforcing the cement slurry with synthetic fibres.

Regarding casing expansion-induced cracking, the consequences of the radial expansion of the steel casing resulting from internal pressure increments were analysed in this thesis. A test setup was designed and built to mechanically expand the steel casing while the cement was monitored and cracks in the cement ranging from 10 to 500 μm width were quantified. While each specimen exhibited a unique cracking pattern without a discernible trend in the measured crack widths, an analysis of the crack areas revealed a consistent trend observed across all specimens. The main output was that the cracked area in the specimens showed i) rapid growth at casing radial expansions between 0 to 100 μm, reaching cracked area values around 15 mm², ii) a gradually slower increase at casing radial expansions between 100 and 250 μm reaching cracked areas up to 25 mm², and iii) a relatively constant cracked area stabilizing at approximately 25 mm² beyond radial expansions of 250 μm.

Once cracking in oil well cement sheaths was characterized, leakage scenarios that can lead to sustained casing pressure (SCP) were evaluated. The literature reports permeability increments in cement sheaths from 10^-19 m2 when uncracked to 10^-12 m2 when cracked and leak rates as high as 10,000 ml/min for gas leaks and 1,000 ml/min for oil leaks. Analytical equations commonly employed to estimate flows and permeabilities tend to overestimate actual flow rates through cracks, primarily due to the omission of key factors such as crack tortuosity, surface roughness and self-healing processes. An experimental procedure to systematically measure flows, determine permeabilities and evaluate self-healing processes in deliberately cracked cement specimens was designed and built to quantify the influence of these factors and define empirical flow reduction factors.

In summary, the presented experiments, testing methodologies and designed evaluation equipment aimed to develop more accurate models and simulations for cement sheaths in relevant oil well conditions based on experimental data.