This research work presents a study of Top of the Line Corrosion (TLC) on carbon steels in the presence of carbon dioxide and acetic acid. The influence of different parameters such as the presence of mono-ethylene glycol (MEG) and the use of pH control were studied in a 4 diameter flow loop. Two sets of experiments were conducted; one at 70ºC, high CO2 partial pressure and a critical condensation rate and another one at 80ºC, low CO2 partial pressure and a high condensation rate. Weight loss techniques and surface analysis were used to evaluate the corrosion rate and products. It was found that the presence of HAc at the concentrations evaluated does not affect the general corrosion rate at the top of the line. Top of the line corrosion rates correlate with the De Waard / Lotz estimate of 10% of bottom of the line rates15. It is believed that the corrosion mechanism is still controlled by CO2 partial pressure at the experimental conditions evaluated in this study. The presence of MEG has not shown any effect on TLC due to the fact that the condensation rate was kept constant. The use of pH control in the supply is believed to limit the amount of HAc in the condensed water. No clear evidence of localized corrosion or pits can be reported at this stage of the study due to the short time of exposure.
TLC occurs when, in multiphase stratified flow or during wet gas transportation, water vapor condenses on the internal walls of the pipeline due to the heat exchange occurring between the pipeline and its surroundings (e.g., offshore or arctic production). Water vapor condenses on the colder walls, forming a thin film of liquid which is enriched in aggressive species, such as organic acids and carbonic acid which comes from the dissolution of carbon dioxide (see Figure 1 ). Gunaltun1 reported that the corroded pipe wall can be divided into three parts:
- The bottom of the pipe, where the corrosion is uniform and where the corrosion rate can be lowered with the use of inhibitors.
- The sidewall of the pipe, where the condensed water drains to the bottom. The corrosion is also uniform, but inhibitors are not efficient.
- The top of the line, where a protective iron carbonate layer can be formed in certain cases, inhibitors are not effective, and localized corrosion occurs.
The main parameters influencing the TLC, as described in many papers2-7, are the bulk
temperature, partial pressure of carbon dioxide, gas velocity, and condensation rate. All of these parameters influence the corrosion rate in a complex way, but a common behavior has been reported:
? At low condensation rates, a protective film of iron carbonate appears at the surface of the pipe exposed to TLC, which lowers the corrosion rate. The formation of this film is explained by the iron saturation of the liquid film and the precipitation of corrosion product.
? At high condensation rates, saturation cannot be reached, and the corrosion rate can be on the order of several mm/y. The corrosion rate is governed by the rate of the corrosive reaction and the rate of condensation. The former increases the amount of iron in the liquid film, while the latter decreases it. The corrosion rate depends on the balance of these two counteracting effects. The corrosion rate is governed by the rate of the corrosive reaction and the rate of condensation. The former increases the amount of iron in the liquid film, while the latter decreases it. The corrosion rate depends on the balance of these two counteracting effects. In 2002, Vitse8 completed a study on TLC in presence of carbon dioxide in the same experimental loop used in this work. He presented several new observations regarding the influence of some parame