Abstract
A series of experiments was conducted in simulated sour service environments comprising brine saturated with CO2, H2S/CO2 mixture or H2S at elevated temperature. In one test a supersaturation of FeS was added to the test environment. A real-time, field/plant corrosion monitoring apparatus was used with multi-technique capabilities incorporating LPR, HDA and EN techniques. The findings showed that the determination of accurate LPR corrosion rates required the use of actual B value measurements determined by the HDA technique. With such an approach, measured electrochemical corrosion rates were within 20 percent of mass loss corrosion rates and often within 5 percent even in highly sour environments. It was concluded that the lack of B value measurement in commonly available LPR devices results in measurement errors of from 0.5 to 3 times and in some cases up to 400 times actual mass loss corrosion rate. In the presence of severe FeS supersaturation, B value correction greatly improved LPR corrosion rate measurement but did not result in measurements that matched the mass loss corrosion rates. In this case, EN corrosion rates correlated best with the mass loss corrosion data obtained by post-test evaluation of the probe elements. Additional functionality of this advanced device also allowed understanding of corrosion spikes by assessment of the galvanic currents between electrodes caused by removal of corrosion films by turbulent action during the test.
