A model has been developed that can predict the corrosion rate in horizontal slug flows. The effect of the slug frequency and oil type on corrosion rate have been included. The model has been compared to experimental data and, to the model and field data of Gunaltun (1996). For all conditions, the corrosion rate increased with increase in slug frequency until a maximum in corrosion rate is reached at approximately 35 slugs/minute.
At 60 C, the model compares well with that of Gunaltun (1996) if a slug frequency of 10 to 12 is used. For 80 C, the Gunaltun model is in good agreement if a frequency of 1 slugs/minute is used. His model does include a term that predicts a maximum in the corrosion rate between 60 and 80 C. This has not been noticed in this laboratory for slug flows.
For horizontal pipelines, field data suggests that, the slug frequency is usually in the range of 1 to 20 slugs/minute, depending on the liquid velocity. When the pipe is inclined, the slug frequency can increase to values much greater than these and this may lead to higher levels of corrosion.
The oil type is accounted for using the suggestion of Efird (1989) based on the product of oil acid number and % nitrogen. When this relation is used, the results compare very well with those of Efird for the oils he studied.
In remote areas such as Alaska or subsea, it is usually not economical to separate locally the oil/water/gas from each well. Consequently, the oil/water/gas mixture from several wells is transported to a central gathering station through large diameter pipelines, where separation takes place.
Levels of carbon dioxide and brine in mature wells can be as high as 30 and 90% respectively. This multiphase oil-water-gas mixture forms weak carbonic acid which can cause much higher corrosion rates inside carbon steel pipelines. The oil and gas mixture may also contain additional components e.g. waxes, hydrates, hydrogen sulphide and sand.
Predictive models for the corrosion rate have been suggested by several workers.. De Waard and Milliams (1975)1 determined corrosion rates by means of weight loss coupons and polarization resistance measurements in stirred beakers They found that the corrosion rate increased with increase in carbon dioxide partial pressure and initially increases with increase in temperature from 30 to 60 C, reaches a maximum between 60 to 70 C and thereafter decreases until 90 C, Vuppu and Jepson (1994)2 obtained similar results from experiments in flow loops under full pipe (oil/water) flow conditions.
Later, de Waadr, Lotz and Milliams (1991)3 provided an improved model which included correction factors for the non-ideality of carbon dioxide at high pressures, formation of iron carbonate scales at high temperatures and changes in pH and Fe2+ ion levels. Further, they (1993)4 produced a revised correlation that included flow velocity. From a limited set of data obtained from a high pressure test facility.
