The Effect of Surfactants on Gas-Liquid Pipe Flows

Abstract

Liquid loading is a major problem in the natural gas industry, in which gas production is limited by the accumulation of liquids in the well tubing. Liquid loading can be prevented by the injection of surfactants at the bottom of the well. The surfactants cause the liquid in the well to foam, thereby changing the gas-liquid flow in the well. The flow is characterized by the TPC (Tubing Performance Curve), which relates the average pressure gradient in the tubing to the gas flow rate. This work has two main goals: (i) To improve the understanding of the effect of surfactants on gas-liquid flow in pipes, which we characterize by a change in the generalised TPC. The generalised TPC relates the average pressure gradient to the gas and liquid flow rates in the pipe. (ii) To provide subsidies for the development of simple physically-based models for the effect of surfactants on gas-liquid flow. We performed experiments in intermediate-scale pipes (lengths of 12 m to 18 m and diameters of 34 mm, 50 mm, and 80 mm) with air and water at atmospheric conditions, without and with surfactants. Multiple parameters, that also vary between different gas wells in the field, were varied: the gas and liquid flow rates, the pipe diameter, the pipe inclination, the surfactant type and the surfactant concentration. We performed a visualisation of the flow without and with surfactants to obtain qualitative results on the effect of surfactants on the flow morphology, and we related these results to quantitative measurements of the generalised TPC and the liquid holdup. The behaviour of the generalised TPC is to a large extent determined by the transition between annular flow and churn flow. In annular flow without surfactants, at large gas flow rates, the water is present in a film along the pipe wall and in entrained droplets in the gas core; the water always moves upwards, which leads to a relatively regular flow morphology. In the churn flow regime, which occurs at low gas flow rates, the liquid film reverses, as the interfacial friction between the gas and the liquid, which drags the liquid upwards, no longer exceeds the gravitational force on the film. This leads to a complex flow morphology, a large liquid holdup and a large pressure gradient. Surfactants cause the formation of foam through the hydrodynamics of the flow. The foam decreases the density and increases the volume of the film at the wall. This changes the balance between the interfacial friction and the gravitational force, which shifts the transition between churn flow and annular flow to lower gas flow rates. As a result, the generalised TPC is changed by the surfactants, leading to a decrease in the pressure gradient at low gas flow rates. An optimum surfactant concentration exists that results in the largest reduction of the pressure gradient. This concentration increases with increasing film thickness; therefore, it increases with decreasing gas flow rate, increasing liquid flow rate, increasing pipe diameter, and decreasing inclination from horizontal. Qualitatively, these results are unaffected by the type of surfactant that is used. From the results obtained in this work, we qualitatively understand the effect of surfactants on the gas-liquid flow, and we understand why surfactants are able to deliquify gas wells. However, a physically-based model is required to translate the results obtained in this work in a quantitative way to the large-scale gas wells. Such a model requires a characterization of the foaming behaviour of the surfactant-liquid mixture using a small-scale setup. We determined that a small-scale sparging setup, often used in the gas industry, is not suitable, because the hydrodynamics in the sparging setup differ too much from the hydrodynamics of annular flow and churn flow. A small-scale shaking test, in which the hydrodynamics more closely resemble churn flow, shows more potential to characterize the foaming behaviour of the surfactants in the context of gas-liquid flows.