Predicting Wax Appearance Temperature and Precipitation Profile of Normal Alkane Systems: A Selective Co-Crystal Model

Wax deposition, occurring when temperature drops or composition changes, has been a great concern in petroleum industry. The precipitated wax usually deposits on the inner surface of pipelines or operating units, and causes severe problems in petroleum production, transportation, and storage. In order to reduce economic loss due to remediation, it is helpful to have an accurate thermodynamic model for predicting wax appearance temperature (WAT) as the onset temperature of wax deposit as well as the precipitation profile. The present work proposes an innovative thermodynamic model, so called Selective Co-Crystal model, capable of predicting the condensed phase behavior of n-alkane mixtures. Based on a systematic literature review of experimental findings, the model is derived by assuming co-crystal formation at solid-liquid equilibrium of binary, ternary and multicomponent n-alkane mixtures. The model selectivity falls into two aspects. One is to determine what chemical species appear in both solid and liquid phases, and the other is the solid phase morphology, either rotator or nonrotating state. The model input only requires pure component properties, i.e. melting temperature, solid transition temperature, heat of fusion, and heat of solid transition. The modified UNIFAC activity coefficient model is employed for accounting the liquid phase nonideality with excellent performance. The model is validated by comparing the predicted WAT with available experimental data of binary, ternary, and multicomponent n-alkane mixtures. Our model predicted results present well the solid-liquid equilibrium in terms of temperature and compositions. In a further step, our model is applied to predict the precipitation profile of synthetic crude oil samples. For these complicated n-alkane mixtures, the equilibrium solid phase is assumed to contain no more than three components. As temperature drops, solid phases are assumed to precipitate out and stack up on one another to form a lamellar structure. With the algorithm incorporated in Visual Basic Application, the model predicted precipitation profile is in good agreement with experimental measurement. In addition, reasonable phase compositions are obtained.