Well Integrity Design for CO2 Injection Wells: Cement, Corrosion and Monitoring

Most CO2 storage wells are drilled and completed by people whose instincts come from hydrocarbon work, and most of those instincts still hold. The exceptions are the design life and the cost of failure. A producer gets abandoned when the field runs out. A leak from a producer is a loss of containment on a well with a finite life. An injector has to hold back a dense, reactive fluid for as long as the storage licence says so. A leak there means losing the stored inventory, and the regulatory and liability consequences outlast the project.

Most of the money goes to three things: remediating leaking or degraded wells, conservative materials selection that the actual service conditions did not justify, and monitoring programmes that produce data nobody can act on. This article covers the three areas where drilling and well engineers moving into CCS most often need to recalibrate: cement barrier performance, tubular corrosion risk, and monitoring.

Standards and requirements context

No single document covers everything. Practical CCS well design draws on several frameworks:

• ISO 27914 covers geological storage of CO2 as part of the carbon capture, transportation and storage standards family. It sets project-level expectations for site characterisation, risk management and monitoring, and well integrity sits underneath it.
• ISO 16530-1 defines well integrity life-cycle governance. Its structure (design, construction, operation, intervention, abandonment) maps directly onto an injector's life.
• NORSOK D-010 sets well integrity requirements for drilling and well operations, including well barrier envelopes and barrier element acceptance criteria. It is a common reference for the barrier philosophy of CO2 wells on the Norwegian shelf and elsewhere.
• In the United States, EPA Class VI permitting is the regulatory route for geologic sequestration wells, with its own construction, testing and monitoring conditions.

A 2024 SINTEF report for Standard Norge (Report No. 2024:00065, CCS Well Design Requirements) addresses where existing well standards need CCS-specific requirements. Read it alongside the NORSOK document if you are working in that regulatory environment. The NETL report Well Integrity in CO2 Storage Operations: Current Understanding and Open Questions (2016) remains a useful summary of what field and laboratory evidence does and does not show. The SPE paper on technical and regulatory considerations for CO2 injection wells (SPE 125839-MS, 2010) is worth reading for how regulators framed the problem early on.

Why CO2 changes the barrier problem

Three features of CO2 service drive the design differences:

1. Chemistry. CO2 dissolved in water forms carbonic acid. Dry CO2 is far less aggressive than wet CO2, so the water phase is the real hazard, not the CO2 itself.
2. Phase and temperature behaviour. Dense-phase or supercritical CO2 changes density with pressure and temperature along the wellbore. Expansion and depressurisation can cause strong cooling. Thermal stresses on tubing, cement and seals during startup, shut-in and blowdown can matter as much as the chemical environment.
3. Time. Barriers have to perform over a storage horizon, not a production life.

Cement barrier performance under CO2 exposure

Degradation chemistry

Portland-based cement is chemically unstable in CO2-saturated water. The sequence reported in the literature, including the comprehensive review in Energy Geoscience (2025), runs as follows:

• Carbonic acid reacts with calcium hydroxide and calcium silicate hydrate phases to form calcium carbonate. Early on, this can reduce porosity and permeability, which looks like self-sealing.
• With continued exposure to CO2-bearing water, the carbonate can dissolve as bicarbonate, leaching calcium and leaving a more porous, weaker matrix.

Which of these dominates depends on water availability, flow, temperature and cement composition. Stagnant, low-water conditions tend to be benign. Flowing, CO2-rich water along a leak path is not.

Where leaks actually occur

Bulk cement attack is slow. The more likely path is a defect in the as-built barrier: channelling from poor mud removal, or a microannulus at the cement–casing or cement–formation interface. Pressure and thermal cycling from injection, shut-in and intervention can open these interfaces. The NETL report puts this emphasis on interfaces and defects, and the SPE 208738-MS paper on CCS well design approaches the problem through load cases rather than chemistry alone. For a drilling engineer, that means:

• Cement placement quality is the first-order control. Hole cleaning, centralisation, mud removal and displacement design matter more than the cement chemistry label on the slurry.
• Design for load cycles. Run casing and cement stress analysis with realistic cold-injection and shut-in transients, not just burst and collapse at steady state.
• Justify the slurry with laboratory testing under representative CO2 exposure. Don't assume it from the datasheet. Options include modified Portland systems and CO2-resistant alternatives. Treat each one as a decision qualified for this well, not a catalogue pick.
• Set the cement evaluation baseline at construction. Without a good baseline log, later logs can't show change.

Converted wells

On repurposed hydrocarbon wells the problem is simple to state: the original cement was designed for a different service, and the record may be incomplete. Treat the checklist below as the minimum for any conversion.

Corrosion risk to tubulars

Mechanism and control

Corrosion in CO2 service is a wet-CO2 problem. The practical question is whether free water will ever touch steel, and under what conditions. Sources of free water include:

• insufficient dehydration of the injection stream;
• brine backflow or crossflow during shut-in;
• condensation after cooling;
• annulus fluid after a tubing or packer leak.

Impurities in the CO2 stream raise the risk further. Depending on the capture source, sulphur and nitrogen oxides, oxygen and hydrogen sulphide can make wet conditions much more aggressive and change which materials are acceptable. Stream specification is therefore a well integrity input, not just a pipeline concern.

Materials choice

Goldeneye shows the design logic clearly. The tubing is the sacrificial or protected element, the annulus is managed, and the casing is protected by keeping CO2 away from it. That only works if tubing and connection integrity is maintained and monitored.

Seals, elastomers and connections

Elastomers can suffer from CO2 permeation and rapid decompression damage, so qualify seals for the actual fluid and for pressure cycling. Use gas-tight premium connections. Check the low-temperature behaviour of metals and seals against the cold transients that expansion can create at the wellhead and upper completion.

Annulus management

Annulus fluid choice and corrosion inhibition decide how the casing fares if tubing integrity is lost. Document what the annulus will contain, what happens if CO2 gets into it, and how you'll detect that.

Monitoring approaches

Monitoring is how the barrier design turns into evidence. Pick each method for what it detects and the decision it supports.

The strongest programmes treat injection stream quality and annulus pressure as continuous leading indicators, and use logging as periodic confirmation. Trend annulus pressures against the commissioning baseline and chase any sustained drift. Don't rely on a fixed alarm number copied from a hydrocarbon well.

Safety during annulus and wellhead work

If diagnostics involve bleeding an annulus or opening wellhead equipment, treat the CO2 inventory as an asphyxiation and cold-injury hazard as well as a pressure hazard. Before work:

1. Isolate the section using verified barriers.
2. Depressurise through a controlled route to a safe vent or collection point.
3. Verify zero energy by measuring pressure at the work location, and confirm there is no trapped pressure between valves.
4. Apply lock-out/tag-out to all energy sources.
5. Set up hazardous-area and confined-space precautions. CO2 is heavier than air and collects in low spots and cellars.
6. Use gas detection for CO2, and for hydrocarbons or H2S if the well history includes them.
7. Vent only to a safe location, controlling the release rate to limit cooling and hydrate or ice formation.

Illustrative scenario: converting an existing well

The following is illustrative and not drawn from a specific project.

An operator is considering a former producer for CO2 injection. The well has carbon steel casing, an older cement job and a record with gaps. A sensible workflow looks like this:

1. Collect the as-built records and offset well data, and identify what is unknown, such as the cement top and the quality of bonding across the caprock.
2. Run cement evaluation and casing inspection logs. Because no earlier baseline exists, treat these as the new baseline and interpret them conservatively.
3. Check the injection stream specification and decide whether a corrosion-resistant tubing string is justified given the credible water and impurity exposure.
4. Run thermal and pressure load cases through casing, cement and tubing, including cold startup and shut-in.
5. If the cement across the caprock is doubtful, choose between remediation (squeeze or section milling and recementing) and a new well. Make that call before committing to the conversion, because failing later costs far more.
6. Set up annulus and injection stream monitoring, and define the response actions in advance.

The outcome depends on the logs. If they show good bonding and sound tubulars, conversion with tubing replacement may be justified. If they don't, a new well is likely the lower-risk path.

Practical checklist

• Confirm the injection stream specification, including water content and impurities, and tie it to the materials selection.
• Identify every credible path for free water to reach steel, including shut-in and backflow.
• Select tubing and connection metallurgy against the real stream, not the nominal one. Qualify elastomers for CO2 and for decompression.
• Run casing and cement stress analysis with cold-injection and shut-in transients.
• Qualify the cement slurry with laboratory testing under CO2 exposure, and specify mud removal and centralisation targets for the job.
• Record a cement evaluation and casing inspection baseline at construction.
• Define the barrier envelopes and acceptance criteria in line with NORSOK D-010 or the applicable regulatory framework.
• Install continuous tubing and annulus pressure monitoring, and define the response to sustained drift.
• Decide which of downhole gauges, fibre and periodic logging each well needs, and what decision each one supports.
• For conversions, document unknowns in the well record and treat them as risks until logging resolves them.
• Write the annulus intervention procedure with the isolation, depressurisation, zero-energy verification, LOTO, gas detection and venting steps.

Conclusion

CO2 well integrity is conventional well engineering stretched over a longer horizon with a reactive fluid. Cement placement quality and interface integrity matter more than bulk chemistry. Corrosion control depends mostly on keeping free water away from steel and on knowing the real stream composition. Monitoring should lead with stream quality and annulus pressure, with logging there to confirm.

Next, take a well you know well and run it through the checklist above, starting with the injection stream specification and the cement baseline. Then read the SINTEF CCS Well Design Requirements report and the NETL open-questions report, and see where your design assumptions rest on evidence and where they rest on practice carried over from hydrocarbon wells.