The acid gas removal unit at GASCO’s Habshan 5: simulation and comparison with field data

The most commonly used process for acid gas capture from natural gas is absorption by means of chemical solvents, such as alkanolamines. This technology is consolidated and widely applied in many industrial plants. Acid gas removal with amines is accomplished in an absorption column, whose design is still difficult, due to the different phenomena involved. Commercially, several process simulators can be found: they are based on different theories and assumptions, both for thermodynamics and for mass transfer with reaction. However, a comparison of the results obtained with the different tools shows significant differences and little agreement with experimental data. Engineering companies require a reliable simulation tool, able to reduce uncertainties in plant design and enabling the user to fulfill the stringent contractual guarantees. In this work a previously developed simulation routine implemented in ASPEN Plus®, tested on pilot plant data, has been applied to simulate the amine section of a huge petrochemical industrial plant recently built in the United Arab Emirates, taken as industrial benchmark. The considered natural gas sweetening unit is part of the Habshan 5 plant, one of the four desert plants of the GASCO’s industrial complex. This is a huge petrochemical site built to deliver a range of marketable products such as methane and heavier compounds and granulate sulfur, sold to customers all over the world. The overall plant capacity is very high (2.2 MMSCFD of gas), making Habshan 5 one of the largest process plant in the world. The natural gas fed to this plant is characterized by the presence of aromatic and sulfur compounds (other than H2S), that can influence the absorption phenomenon, making the multiphysic approach adopted for the model even more complicated. The comparison between simulation results and experimental field data shows a good agreement, proving the reliability of the developed simulation tool.