Abstract
As oil fields mature, operating costs associated with internal corrosion of oil transport pipelines are increasing. Beside aging infrastructure, this trend is caused by decreasing oil production and increasing co-production of water. Even if water is heavier than oil, it can be entrained in the oil phase if the flow of oil is sufficient for the turbulent forces to overcome the density difference. Determining the minimum velocity for which this is possible (the entrainment velocity) is an important factor for establishing good corrosion management strategies. To determine this velocity, a steel surface wettability study is conducted on a model oil blend (43% paraffinic, 26% isoparaffinic, 27% naphthenic and 4% aromatic) with physical properties that mimic those of an actual crude oil (API0 37.50, viscosity 7.3 cP at 25.60C, interfacial tension of 39.3 mN/m and a contact angle of 1310). Using a doughnut cell benchtop apparatus, it was found that the wetting behavior of the crude oil could not be mimicked with a single model oil. Rather, a blend of model oils that more closely resemble the crude oil chemistry is needed to accurately represent the wetting properties of the crude oil. The empirical data was compared to water wetting prediction models developed by Hinze 1955, Tang 2011 and Cai 2012. The Tang model was found to give the closest prediction of the empirical data. Such models can be utilized for the design and sizing of pipelines to avoid water wetting, stratified flow and subsequent corrosion. Moreover, they can be used as a tool to adjust corrosion inhibition plans in terms of dosage rate and type of chemicals used.
