Abstract
Foamed cement has provided zonal isolation in oil and gas wells since the 1980s. However, the lack of an experimental technique that fully characterizes foamed cement properties is limiting its application. This study uses X-ray microcomputed tomography (micro-CT) to elucidate relationships between the foaming process, foamed cement microstructure or morphology, and macroscopic performance of the material’s mechanical properties.
During this study, foamed cement slurries were prepared using a traditional multiblade laboratory blender to investigate the influencing factors on foamed cement properties. The influences of shear rate, mixing energy, surfactant concentration, and base cement slurry composition on the properties of set foamed cement were specifically studied. A MACS® multiple analysis cement system was also used to study the gas pressure and blender geometry effect on test results. Approximately 25 foamed cement slurries with foam qualities (FQs) ranging from 20 to 80% were produced. The microstructure and macroscopic performance of the foamed cement were quantified using micro-CT analysis and uniaxial compression tests.
Test results indicate there is an important mixing energy threshold value that needs to be supplied to produce stable foamed cement. The maximum achievable FQ is determined by the shear rate during the foaming process. Once the minimum qualifications for obtaining stable foamed cement and a target FQ are met, a further increase in mixing energy or shear rate has little effect on the foamed cement microstructure or macroscopic performance. However, excessive mixing energy supplied at a high shear rate can lead to a higher initial slurry temperature, a less homogenous microstructure, and relatively poor mechanical properties after the foamed cement has set. The primary determining factors of the foamed cement microstructure are FQ and gas pressure. In samples generated using the multiblade laboratory blender at atmospheric conditions, the median gas bubble size increased significantly from 116 µm at 20% FQ to 1400 µm at 80% FQ. The median gas bubble size in the 20% FQ samples generated using the MACS device decreased significantly when gas pressure increased from 133 µm at atmospheric pressure to 39 µm at 1,000 psig.
