Sweet Corrosion

Key Takeaways

• CO2 corrosion rate prediction uses the de Waard and Milliams model (1975) as the foundational correlation, which expressed the CO2 corrosion rate of carbon steel as a function of temperature, CO2 partial pressure (pCO2), and wall shear stress: log(CR) = 5.8 - 1710/T + 0.67 * log(pCO2) (simplified form, in mm/year, T in Kelvin), indicating that corrosion rate increases with temperature (faster kinetics) and with CO2 partial pressure; the model has been extended and revised by de Waard, Lotz, and Milliams over subsequent decades to incorporate the effect of iron carbonate film formation (which reduces the corrosion rate when protective scales form) and the effect of organic acids (acetic acid, propionic acid) present in some gas condensate wells that significantly accelerate CO2 corrosion by acting as additional proton donors beyond carbonic acid; modern CO2 corrosion prediction software (NORSOK M-506, ECE, Cassette by IFE) uses mechanistic models that account for film formation, flow regime, steel microstructure, and water chemistry to predict the corrosion rate distribution along the tubing string from the perforations to the wellhead under the pressure and temperature conditions of each specific well.
• Protective iron carbonate (FeCO3) scale formation is the key variable determining whether CO2 corrosion is manageable or catastrophically fast: FeCO3 precipitates when the product of the iron ion concentration and the carbonate ion concentration exceeds the solubility product (Ksp) of siderite; at higher temperatures (above 60 to 80 degrees Celsius) and with sufficient iron available from the dissolving metal surface, FeCO3 can form a dense, adherent, low-porosity scale layer that acts as a diffusion barrier between the corrosive solution and the steel surface, reducing the corrosion rate by factors of 10 to 100 compared to the bare-steel rate; at lower temperatures (below 40 to 60 degrees Celsius) or in high-flow-velocity conditions where shear stress removes the partially precipitated scale before it densifies, FeCO3 forms a loose, porous, non-protective film (or fails to form at all), leaving the steel surface exposed to the full corrosive attack; the transition temperature for protective FeCO3 film formation is itself a function of the CO2 partial pressure, water composition, and flow velocity, and must be evaluated for each well's specific conditions; the infamous mesa attack or table-top corrosion morphology occurs where FeCO3 forms a protective film in some areas but the film is locally undermined and removed by flow turbulence or local flow disturbances, leaving isolated mesas of original steel surface surrounded by heavily corroded valleys.
• Material selection for CO2-corrosive service is the most reliable long-term mitigation for sweet corrosion in high-pCO2 wells: the Corrosion Resistant Alloys (CRAs) typically used in sweet corrosion service include 13 Cr stainless steel (martensitic stainless with 13 percent chromium, providing significantly better CO2 corrosion resistance than carbon steel through the formation of a stable chromium oxide passive film), 22Cr duplex stainless steel (a two-phase ferritic-austenitic microstructure with higher strength and better corrosion resistance than 13 Cr), 25Cr super duplex (for the most aggressive CO2 and chloride environments), and nickel alloys (Alloy 825, Alloy 625) for extreme conditions; 13 Cr is the most commonly specified CRA for gas condensate wells with moderate to high CO2 partial pressures, providing adequate corrosion resistance at costs 2 to 4 times the equivalent carbon steel tubular; material selection charts (plotting CO2 partial pressure versus temperature versus chloride content versus H2S concentration) guide the selection between carbon steel (with chemical inhibition), 13 Cr, 22 Cr duplex, and higher alloys based on the severity of the corrosion environment; the critical parameter is the CO2 partial pressure at the bottom of the well (where temperature and pressure are highest) and at the wellhead (where temperature has dropped and condensation creates a separate liquid water phase that is particularly corrosive).
• Corrosion inhibitor injection is the primary mitigation method for sweet corrosion in carbon steel tubing when material upgrade is not economically justified: film-forming corrosion inhibitors (FFCI, typically imidazoline, amine, or quaternary ammonium compounds dissolved in a carrier solvent) adsorb on the steel surface to form a hydrophobic monolayer that displaces the corrosive water phase from the steel surface and reduces the corrosion rate by 90 to 99 percent when sufficient inhibitor concentration is maintained; continuous injection of corrosion inhibitor through a downhole chemical injection mandrel at the required treat rate (typically 50 to 300 ppm of inhibitor in the produced water phase) maintains inhibitor concentration above the minimum effective concentration throughout the tubing string; batch treatments (pumping a concentrated inhibitor slug followed by a carrier fluid to push the slug to the bottom of the well) are used in wells without continuous injection capability, with batch frequency (daily, weekly, monthly) determined by the inhibitor film persistence (the film durability under flowing conditions); inhibitor evaluation includes rotating cage tests and flow loop tests under simulated downhole conditions to confirm the inhibitor's effectiveness before deployment, and field verification by corrosion coupons (weighed metal samples installed in the flow stream and retrieved periodically to measure weight loss as an indicator of corrosion rate).
• Process modification to reduce CO2 partial pressure (pCO2) is a prevention strategy applicable at the surface facility level: CO2 removal from the produced gas stream by amine absorption (MEA, DEA, MDEA gas treating units) reduces the CO2 content and therefore the partial pressure of CO2 in the reservoir and tubing; injection pressure management (maintaining higher wellbore pressure to reduce the CO2 liberation from the produced water) reduces the severity of sweet corrosion in the wellbore; dehydration of the produced gas (glycol dehydration to remove water vapor) eliminates the aqueous phase in the surface piping and flow lines where CO2 corrosion in the gas phase requires liquid water to be present; in some gas condensate fields, the CO2 content of the gas increases over the producing life as CO2-rich formation water is increasingly produced from deeper or more CO2-saturated intervals, requiring a reassessment of the corrosion mitigation strategy as the CO2 partial pressure and water cut evolve during field depletion.