Ensuring well integrity during CO2 injection.

Wells have in numerous scientific publications been denoted the "weak link" of safe and cost-efficient CO2 Capture and Storage (CCS). Whether they are active or abandoned, these wells are all man made intrusions into the storage reservoir and their sealing abilities depend on degradable materials like steel and cement. To ensure that stored CO2 remains underground in a long-term perspective, it is necessary to advance current well technologies, procedures and materials. As opposed to normal petroleum wells, wells penetrating CO2 storage sites can be exposed to very low temperatures. This can be the case e.g. during injection, or if a leak develops in the well. A prerequisite for maintaining well integrity during CCS is therefore to understand how wells can be constructed (tailor-built or modified) to withstand low temperatures. Especially the strong temperature variations are a concern, as it is well-known from the petroleum industry that thermal cycling can have detrimental effects on well integrity. Repeated heating and cooling of the well will cause materials to expand and contract, and especially the annular sealant material (typically cement) is likely to de-bond and crack radially. This will create leak paths for formation fluids. Before the start of this project, few experimental studies had been made to understand the effect of cyclic temperature variations on well integrity and no studies had covered the temperature interval relevant for CO2 injection wells. The goal of the current project was therefore to try to close this knowledge gap. The project team has consisted of drilling- and well researchers from SINTEF Petroleum collaborating with gas technology scientists from SINTEF Energy and geochemistry/geomechanics experts from Lawrence Livermore National Laboratory (USA). We have collaborated to understand more about when, where, how and why well leaks develop during cyclic temperature variations and how such detrimental cycles come about. The project has had a duration of 3 years, and its focus has been threefold: (1) Experiments have been made to study how well cement adheres to steel and rock, (2) Experiments have been performed to study how downscaled well samples resist temperature variations, (3) Models have been developed based on the experiments to give information about how to best construct and operate CO2 injection wells. The major findings of the projects are: (i) New experimental methods to characterize and measure interface strength between well cement and steel/rock. These tests all show that drilling fluids have a strong impact on bonding quality, and should be removed/chosen carefully before cementing. (ii) Wells are relatively robust towards temperature variations as long as the temperature is above the freezing point of pore fluid in cement/rock. (iii) We have developed a new experimental-numerical method to estimate leakage through or along well cement. It shows that even small defects in the cement (if connected) can cause high permeability. (iv) A set of models has been (further) developed to enable modelling of vertical CO2 flow in wells including heat transfer with the surroundings. The models have been used to study various challenging situations/scenarios, e.g. cold CO2 injection, shut-in/restart of wells and the dynamics of a potential CO2 blow-out. Based on the results we provide industry with research-based advice on how to avoid leakage in CO2 injection wells by focusing on material choices, well design and operational parameters.

A concern in Norway when it comes to large-scale CO2 Capture and Storage (CCS), or the application of CO2 injection for Enhanced Oil Recovery (EOR), is that Norwegian CO2 point sources are few, geographically spread and relatively small. As a result, ship transport will in many cases be favoured over pipelines. The major drawback with a carrier-based CCS solution is that CO2 needs to be in a dense state for optimal shipping efficiency (i.e. liquid and at -53°C). A first step towards making such a transpo rt and injection scheme possible is to understand how wells can be constructed (tailor-built or modified) to withstand these low temperatures. Injection of fluids with temperatures down towards -50°C will expose the well to large thermal cycles during nor mal operations such as injection/shut-down, and a prerequisite for such injection is that the well materials can withstand these thermal cycles. It is well-known from the petroleum industry that thermal cycles in a well can have detrimental effects on we ll integrity. Especially, the annular sealant material (e.g. cement) is likely to de-bond and/or crack radially, which leads to leak paths for formation fluids. Few experimental studies have been performed on the effect of thermal cycling on well integrit y so far, and no such studies have been performed in the temperature range relevant for CO2 injection. This project aims to study through numerical modelling and experiments when, why, where and how well integrity is lost when a well is repeatedly coole d down and heated up - and how such detrimental downhole temperature cycles arise. The deliverables will be new knowledge about well integrity of CO2 injection wells, as well as specific recommendations on material selection, well design and operational parameters in order to ensure the long-term integrity of CO2 injection wells.