By Robert R. Petrie
The existence of free water, i.e., liquid, as a separate phase in the bottom section of flowing or stagnant crude oil lines is of prime concern. This free water phase provides the electrolyte necessary for the progression of various forms of corrosion, utilizing the corrosive agents most typically found in the crude oil, such as hydrogen sulfide (H2S). Besides the more benign or tolerable “general corrosion” caused by H2S, both sulfide stress corrosion cracking (SSCC) and hydrogen induced cracking (HIC) may also occur, with their potential for leading to unexpected, catastrophic failure of the crude oil line pipe.
The rigorous prediction of whether a free water phase, stagnant or entrained, will form can be made for any given hydrocarbon system, based on hydrocarbon composition, initial water saturation or undersaturation condition of the hydrocarbon, and the lowest expected temperature of the flowing or stagnant system. The subject, typical crude oil considered here, exiting the desalter/dehydrator train at a gas-oil separation plant (GOSP) is theoretically saturated with dissolved water at the exit temperature of the dehydrator vessel. Any cooling of the crude in the exiting pipelines will result in oversaturation of the oil with respect to water content and subsequent free water formation.
Such a calculational exercise is generally considered academic, since free water entrainment from upstream equipment, operational upsets, hydrotesting, etc., all may lead to the existence of significant amounts of free water in the system in question, regardless of the calculation/prediction. In short, free water in some quantity is often simply considered to be present in pipelines carrying desalted crude.
With the assumption that some free water has formed in the desalted crude, the next task is to ensure that this free water remains entrained in a flowing system or becomes re-entrained in either a stagnant or low-flowing system that has allowed this water to drop to the pipe bottom. The entrainment of the free water phase will in turn ensure that the electrolyte/corrosive agent in contact with the line pipe wall is generally avoided, thus precluding, or at least minimizing, the corrosion reactions.
Beginning in the 1950’s, Shell Development Company in the US, investigated theoretically and empirically the entrainment of settled, free water in a pipeline by flowing crude oil1. The impetus for Shell’s work was the same corrosion concern as described above; their prime goal was the prediction of flow conditions which lead to complete entrainment of water by oil, thereby essentially eliminating the probability of corrosion. Shell’s work included the study of the transportation of solids, e.g., sand, by flowing fluids and comparison of the analogous behavior of free water droplets which had been “broken” away from a stagnant water layer by flowing oil2. From that work, a graph and the equations of the forces working on the deposit (sand or water) sitting on the pipeline bottom are shown in the Appendix of this article. The combined results of Shell’s efforts appeared in their 1975 article3 . Shell confirmed that their correlation had been verified by subsequent model tests at their laboratories in Amsterdam; furthermore, the correlation was being used as a standard design/operational tool in Shell organizations worldwide.
The predictive correlation appearing in that article is the basis of the calculations and conclusions presented here specifically for Arab light (AL), Arab medium (AM), and Arab heavy (AH) crudes.
The working equations/graph for the Shell correlation are included in the attached Appendix. We decided to formalize, for a wider audience, its usage for the world’s\ most ubiquitous crude, Arab light. The calculation for the entraining AL flowrate is based on actual measurements of the key physical properties considered in the correlation: crude-water interfacial tension, crude viscosity, and produced water viscosity.
Actual samples of “typical” Arab light crude and water production were obtained from a producing well in Saudi Arabia and the following physical properties, including interfacial tension, measured over a range of temperatures, except as noted:
* extrapolated from 60°F measurement
AL = 34° API (measured)
In order to cover a reasonable range of Saudi Arabian crudes with regard to specific gravity, it was decided to also perform the entrainment calculations for AM and AH crudes. Measured/published API gravities and viscosities for AM and AH were used as benchmark values; gravities/viscosities at the temperatures of interest, 50°, 80° and 110°F, were obtained by extrapolation of these numbers. Interfacial tension values were also determined by extrapolation. The physical properties of the produced water were taken from the AL measurements or extrapolated as noted. Past calculations had indicated that the calculated entrainment flowrate was relatively insensitive to reasonable variations, e.g., inaccuracies, in these physical properties. The recommended design guidelines with respect to entrainment flowrate resulting from this work were intended to address typical variations in crude/water physical properties in any event.
The following properties therefore were used in the calculational procedure presented here:
*extrapolated from 60°F measurement; water from AL production
AM = 31.3° API
*extrapolated from 60°F measurement; water from AL production
AH = 26.8° API
The results of applying the correlation to the three noted Saudi Arabian crudes are contained in Figures 1/2/3 for AL, AM and AH, respectively. Shell’s work with a given crude oil has indicated a typical concave-down curve of crude velocity (ordinate) vs. pipeline inside diameter (abscissa), reflecting the effects of all the involved physical properties of the crude and water. The attached plots for AL and AM have been effectively “smoothed” primarily because of round-off in the calculations. All three crudes do exhibit some degree of concavity, per the noted calculation, and accented in the graph for AH, which exhibits a more pronounced concave shape.
An interpretation of the crude/water physical properties as reflected in the Figures as well as the correlation equations illustrate the following:
For a given crude oil:
It would be prudent to apply a healthy safety factor in determining a minimum, operational crude flowrate which will either a) maintain a flowing water phase in the entrained state, or b) re-entrain settled water from the pipe bottom. A somewhat arbitrary 25% safety margin, addressing the variation in physical properties of the crude/water phases, thus yields the recommended “operational” lines on the Figures.
Alternatively, pipeline operators could conceivably select minimum crude velocities on a seasonal basis for given diameter pipelines.
Minimum crude flowrates have been determined which will maintain production water in the entrained state and avoid bottom settling and its consequent, pipeline corrosion problems. A combined theoretical/empirical correlation, published by Shell, and proven by model testing, was employed for these calculations. The flowrates have been calculated for typical Arab light, Arab medium and Arab heavy crudes, over a range of temperatures and pipeline inside diameters. Measured physical properties, for the most part, for the actual AL crude and production water have been used in the calculation; published/extrapolated data were utilized for the AM and AH oils. Examination of the results allow analyses of the effects of the pertinent variables, as they influence the calculated minimum crude flowrates, and conservative recommendations of minimal operational crude flowrates to obtain the desired water entrainment.