Abstract

ABSTRACI' A review has been made of various corrosion models available for the prediction of the life of carbon steel tubulars in carbon dioxide containing environments. These models range from relatively primitive empirical relationships to more sophisticated methods that involve the liquid film inside a production string and the solubility of iron carbonate. One of the models shows the prospect of being able to predict the onset of erosion corrosion. Particular emphasis has been placed on the corrosion models that are part of the USL Corrosion Computer Model. INTRODUCTION The ability to predict the life .expectancy of a carbon steel tubular in a gas condensate well containing carbon dioxide is a goal of corrosion engineers in the oil and gas production industry. Gas condensate wells are much more complex than typical oil wells since in addition to temperature-pressure changes with depth, there are phase equilibrium changes and flow dynamic conditions that must be considered. The final step of converting this information into a corrosion rate and then a tubing life is perhaps the most difficult task. Two excellent NACE publications (1,2) have clearly illustrated the complexity of carbon dioxide corrosion. A number of investigators have attempted to predict corrosion rate in wells containing carbon dioxide. - Perhaps the best known work is that of DeWaard and Milliams (3), which relates corrosion rate to partial pressure of CO2 and temperature. Others like Crolet and Bonis (4) have developed proprietary models to predict tubing life from field data. The COPRA correlation developed by Gatzke and Hausler (5) is an empirical approach which uses iron counts and production rates of fluids to establish corrosion rates. The Corrosion Research Center at the University of Southwestern Louisiana has developed a computer program which successfully models gas condensate wells using first principles (6,7). Phase III of the program consists of 7 programs listed as Models A-G. The program was released to sponsoring companies during the spring of 1988. This program can be divided into two groups of models that physically describe a gas condensate well: Models A, C, E, and G; and those that predict tubing life: Models B, D, and F. Although Models B and D can be run without any dependency on other models, Model F must have information from other models before it can be run. A brief description of each of the physical models follows:Model A - This is the temperature-pressure profile model which allows one to calculate the temperature and pressure at any point in the production string given the bottom hole temperature and pressure. The calculations are performed for a dry gas and a gas condensate well.Model B - This phase B equilibrium model uses the Peng-Robinson equation of state to calculate the phases that exist at any point in the production tubing. There is a possibility of a single phase, of two-phases, or of three-phases existing in the tubing.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call