At a recent symposium (RGIT 198B6) the Panel Discussion centred around the problem of rate prediction for downhole corrosion in offshore wells. It was generally agreed that methods of rate prediction were inadequate, and that this was in stark contrast to the diagnostic methods that were available using wireline. These diagnostic techniques were considered generally adequate. This contribution attempts to show that a suitable downhole corrosion database would go a long way towards satisfying the corrosion rate prediction requirements of the industry, and would have certain additional advantages.
As the North Sea becomes a mature oil- and gas-producing area, an increasing number of wells are experiencing corrosion and scale problems. Some problem wells are known to have a complete tubing change every 2 years at a material cost per well in excess of di300000. In the interests of simplicity, I will concentrate mainly on tubing problems, but much of the content applies equally well to casing and other downhole equipment
For the corrosion reaction to proceed there must be:
an electrolyte;
an oxidizing agent;
a conductive path in the metal.
In many cases the corrosivity of the electrolyte is difficult to determine-if, for example, a ‘good’ electrolyte such as salt water is emulsified in a ‘poor’ electrolyte such as crude oil.
The driving forces in the corrosion reaction are the oxidizing agents, commonly oxygen, carbon dioxide and hydrogen sulphide. By far the most powerful is oxygen, which remains effective even at very low concentrations. Carbon dioxide becomes significant when the partial pressure exceeds 0.2 bar. Hydrogen sulphide has a similar if higher threshold, but acceptable limits depend on whether general corrosion or stress corrosion cracking is the major problem.
Corrosive effects can produce local pitting or large areas of fairly uniform metal removal, corresponding in some cases to a conductive path in the metal of a few micrometres to several metres. Material homogeneity and flow conditions of the electrolyte are major influences in this respect.
At first glance, the injection wells should be the most easily handled from a corrosion point of view. This is because the quality of the injection water can be carefully controlled It is true that salt water is a ‘good’ electrolyte, but the likely oxidizing agents can be controlled. Oxygen can be stripped, and carbon dioxide and hydrogen sulphide are not normally present in significant quantities. A problem in injection wells is concerned with gas production downhole. The most common variety is hydrogen sulphide produced by sulphate-reducing bacteria (SRB). These bacteria thrive in an oxygen-free environment, and some operators deliberately leave some oxygen in the injection water to control SRB. In this way there is some--hopefully a minimum-amount of corrosion allowed to take place. Stainless steel is only ‘stainless’ in the presence of oxygen, and this can be a further reason for leaving traces of oxygen in injection water.
