Can cement integrity be predicted before it becomes a containment risk?

Once a depleted oil or gas field is selected as a candidate for a possible CO₂ storage project, a first-pass well integrity study is often carried out to assess and rank wells according to the containment risks they may pose to the project. This can quickly identify the most problematic wells, for instance, where a required cement barrier is missing or known to be inadequate.

“No amount of modelling can redeem these obviously problematic wells,” says James Hanson, a researcher at Elemental Energies with a background in physics and computational chemistry. “Instead, we focus on wells whose barriers have passed the standard decommissioning assessment but require further evaluation of the pressure and temperature conditions expected throughout the CO₂ storage project,” James explains, “as well as the potential chemical effects of contact with the CO₂ plume. They may look unassuming at first glance, but they might be harbouring a risk. That is where our modelling work can make a real difference.”

First principles

James teamed up with Maria Chavez and Juan Chavez to build an entirely new modelling approach based on first principles. Together, they form a team that provides critical expertise from three different key disciplines.

“CO₂ injection changes reservoir conditions, causing the cement behind casing and the surrounding annular system that prevents the upward migration of CO₂ to evolve both mechanically and chemically,” says Juan, a geomechanics expert at Elemental Energies. “That is the process we want to model: a dynamic mechanism in which chemistry, geomechanics and fluid flow evolve together.”

“Our drive is to innovate,” says numerical modelling expert Maria, “and we felt that there is a need to do this in the CCS space. What we often come across in the literature is the assumption that leakage through cement is modelled by arbitrarily assuming a fixed leakage pathway. That approach is not sufficient to accurately map what happens in the subsurface, because the formation of a leakage pathway is a dynamic process. That is why we are integrating all these aspects, chemical and mechanical, into one tool.”

Long term

“This will also allow us to model and predict a set of scenarios when it comes to how a well will behave once injection has stopped,” says Maria. “And not only for legacy wells, but maybe even more importantly, for the injection wells themselves. At the end of the day, it is the injection well that will experience the biggest change in reservoir conditions.”

“We have already seen regulatory requirements moving towards the need to better predict leakage risks over hundreds of years of time, following the cessation of injection. Our approach is tailored to addressing that,” James adds. The team is now working hard to finalise the software and plans to bring this numerical modelling work to market as a specialist service.