Improved thermodynamic investigation of asphaltene precipitation

Asphaltenes are analogous to the “cholesterol” of crude oils, so they may cause significant flow assurance problems to various oil and gas processes and negatively affect the economy of the oil recovery, transportation, and processing by increasing operational expenditures (OPEX). Asphaltenes increase oil viscosity, decrease its market value, and, when they precipitate, cause flow assurance challenges. Understanding asphaltene precipitation and phase behaviour is important to avoid, prevent, and address asphaltene flow assurance challenges. An experimental investigation is time-consuming and requires laboratory expertise with limitations on how many experiments can reasonably be conducted over what range of feasible operating conditions. Furthermore, we need to predict asphaltene and fluid phase behaviour over the full range of operating conditions to avoid flow assurance issues. Therefore, having a thorough knowledge of the phenomenon and applying asphaltene modeling approaches is essential to foresee conditions leading to asphaltene precipitation to treat the phenomenon properly. Despite significant research, asphaltene behaviour in different operating conditions and the application of improved thermodynamic investigations have not been well understood. There is little research on the investigation of the operating conditions and improvement of the thermodynamic models (e.g., application of advanced optimization technique) on asphaltene precipitation. This thesis uses different modeling approaches (e.g., equation of state) to investigate crude oil asphaltene precipitation at operating conditions. Asphaltene phase separation can be triggered by altering the operating conditions, e.g., temperature, composition, and adding n-alkanes. For instance, decreasing temperature from reservoir conditions leads to asphaltene precipitation due to alteration of the solubility of asphaltene in the oil mixture. Moreover, the composition of crude oil is upgraded or downgraded by adding different hydrocarbons at the refinery inlet. Yet, the prediction of asphaltene precipitation and the impact of operating conditions are quite uncertain, and detailed thermodynamic investigations and appropriate techniques for adjusting the models are required. Several research studies have used thermodynamic equations of state (EoS) to model asphaltene precipitation. Recently, advanced EoSs that take into account the association of hydrogen bonding has become popular. For example, Cubic Plus Association (CPA) has shown promising results in modeling asphaltene precipitation. There is uncertainty in using EoSs, e.g., tuning the adjustable parameters. Hence, there is a need to systematically study how to adjust the tunable parameters to predict asphaltene precipitation using advanced EoS. The objective of this research is to investigate and improve the performance of EoS modeling of asphaltene precipitation. For this purpose, first, a comprehensive literature review was conducted to address asphaltene precipitation from different standpoints. While a comprehensive literature review to study asphaltene precipitation and deposition was missing in the literature, the focus of this research is to provide an overview of the nature and physical properties of asphaltenes, experimental and thermodynamic/simulation tools investigations, operating/fluid/reservoir impact, inhibition/treatment, and economic analysis of flow assurance. The literature review findings highlighted two main gaps in asphaltene thermodynamic modeling; 1) only gradient-based optimization techniques have been used to tune the EoS parameters, and 2) the effect of heteroatoms in asphaltene precipitation has not been considered. Therefore, the two other objectives of this thesis are tailored to address the gaps. In order to address the fact that only gradient-based methods have been used to tune the parameters, we used a global optimization approach instead of gradient-based optimization to relate and correlate hydrogen bonding to the binary interaction parameters of the Cubic Plus Association (CPA) EoS model. While the application of advanced optimization methods and a systematic sensitivity analysis of operational conditions/BIPs were missing in the literature, the focus of this section is to consider the association of hydrogen bonding in asphaltene precipitation while developing correlations for binary interactions (BIs) using global optimization. The advantages of using global optimization are to avoid entrapment in local minima while optimizing the parameters of the EoS and to improve the correlation/prediction capability of the EoS by finding the best fit of the adjustable parameters. The CPA EoS is validated by predicting unseen data, comparing with cubic EoSs, i.e., SRK and PR, using different oil characterization, e.g., SARA analysis, and drawing an analogy between scaling equation and CPA. Application of the proposed technique significantly improved the performance of the CPA EoS in modeling asphaltene precipitation (average deviation of less than 0.067 for correlation and prediction). The relative importance analysis revealed that the composition of the mixture (dilution ratio) is the most influential factor contributing to the asphaltene precipitation (other factors are temperature and carbon number of the diluents). The effect of polar forces due to the presence of heteroatoms on asphaltene phase behaviour is investigated using a Cubic Plus Polar EoS (CPP). To the best of our knowledge, we have not found any literature focused on polar heteroatom forces in asphaltene thermodynamic modeling. In this novel work, we demonstrate how a single term that accounts for polarity can be added to the extension of the cubic EoS and be effectively applied to calculate asphaltene precipitation. Further, a simplified oil characterization method is adapted to reduce the number of adjustable parameters (binary interactions) and reduce the need for experimental measurements. A global optimization approach and molecular dynamic (MD) simulation have also been used to increase the reliability of the optimization and reduce the number of adjustable parameters for polar forces. This section of the research finds that the CPP approach using global optimization to tune parameters of the EoS is the most reliable approach, followed by CPP EoS using MD to find dipole moment for the aryl-linked core asphaltene structure (average R2 for both modes are above 0.98). The improved thermodynamic approaches (global optimization and including the effect of heteroatoms) introduced in this research can be used by other researchers to increase the efficiency of the asphaltene thermodynamic modeling