A Mechanistic Model for Predicting Top-of-the-Line Corrosion in CO2/H2S Environments: Validation and Field Application

Abstract

This study develops a mechanistic model to predict top-of-the-line corrosion (TLC) in wet natural gas pipelines under sweet (CO₂) and sour (CO₂/H₂S) conditions. The framework incorporates water condensation, CO₂–H₂S–organic acid aqueous speciation, electrochemical kinetics, and corrosion product film formation, emphasizing rapid FeS precipitation in sour service. The proposed sour model reproduced published laboratory data with a mean absolute deviation < 45% across gas compositions ($p_{{\mathrm{CO}}_2}$:2-10 bar, $p_{H_{2}S}$:0-2 bar), wall temperatures (15°C–75°C), condensation rates (0.002–0.25 ml/m²·s), and organic acid concentrations (0–1300 ppm), while the deviation for sweet service was < 16%. The sour TLC model applied to two offshore sour natural gas transmission pipelines showed strong agreement with in-line inspection data. Field simulations indicate that the highest sour TLC rates occur at steel surface temperatures of 20°C–35°C, with a maximum corrosion rate of 0.9 mm/year. High condensation rates alone do not cause TLC; significant corrosion can occur even at low condensation rates (0.00002–0.15 g/m²·s). This work establishes one of the first unified mechanistic framework for predicting sour TLC.