The molecular simulation of oil is employed to understand the behavior of oil in porous structures of oil reservoirs, which make it harder to extract hydrocarbons. The most common model used today is a single-component one, which does not accurately reflect the composition of oil, leading to unreliable results.
The scientists from Skoltech, Moscow State University and Artificial Intelligence Research Institute (AIRI) have attempted to create a more reliable model on the basis of the so-called contact angle, an indicator reflecting the degree of wettability of various oil components by liquid. The authors managed to create an algorithm based on a complex 15-component oil model using empirical data.
In the new algorithm, the reservoir cell has the form of a slit-shaped pore formed by quartz plates, between which oil and water are located. Oil and water in the pore form a drop in the shape of a flattened cylinder within the boundaries of the cell. The projection of a liquid drop has four points of triple-phase contact. The contact angle is calculated as the arithmetic mean of the angles at the four points.
“The new numerical method of calculating the contact angle differs from the existing alternatives due to the linear complexity of determining the angle at each step of the system and the non-necessity of fine-tuning the algorithm for dissolved components, such as methane in water and water in oil. This allows us to speed up the calculations and process large amounts of information in a simpler calculation process,” postgraduate student Petr Khoventhal is quoted as saying by Skoltech.
A key achievement of the study is the fact that it takes into account multiple components of oil, including asphaltenes and methane. This has made it possible to produce a more accurate simulation of molecular interactions.
“Our study has confirmed that asphaltenes are a key oil component in the study of wettability. Methane goes hand in hand with oil and can make up a significant share of its composition, even by mass. Ignoring what could be the dominant component cannot produce accurate simulation results. In a single-component model, these substances are completely ignored, which significantly distorts reality,” Ilya Kopanichuk, one of the authors of the study and candidate of chemical sciences, is quoted as saying by Skoltech.
The new algorithm allowed the scientists to assess the impact of various factors on the contact angle, which decreases with rising temperature and growing concentration of underground brine. In turn, an increase in methane concentration leads to a higher contact angle, whereas the content of aromatic compounds has virtually no effect on this indicator. As a result, this algorithm makes it possible to tailor oil composition based on data from any field.