A modified Zudkevitch and Joffe method of determining parameters of an equation of state (EOS) is described and applied to the Patel and Teja (PT) equation. All the parameters of the PT equation are considered temperature dependent and evaluated from the vapour pressure, liquid molar volume and critical compressibility factor of pure compounds. The attraction term in the PT equation is modified by a temperature dependent parameter for super critical components. Conventional random mixing rules are applied to obtain the EOS parameters for mixtures, with no binary interaction coefficients for hydrocarbon mixtures.
The modified equation was validated against experimental data of gas condensate systems and compared with other leading EOS. A large number of data, including the dew point pressure, condensate drop-out, and the density of gas and condensate phases at equilibrium conditions over a wide phases at equilibrium conditions over a wide temperature range was generated and used in this validation. The proposed method predicted the experimental data, particularly the retrograde condensate volume, more accurately than others. The method can be applied to any equation of state.
Cubic equations of state are commonly used in the petroleum industry for volumetric and phase petroleum industry for volumetric and phase behaviour modelling of reservoir fluids. Since the introduction of van der Waals equation in 1873, a numerous number of equations have been published in the literature and promising new equations keep appearing. In general, the performance of these cubic equations is good for a hydrocarbon system which is predominantly oil. However, it deteriorates for phase behaviour modelling of gas condensates particularly in the retrograde region. particularly in the retrograde region. Most of the equations of state developed so far are primarily to predict the physical properties of pure primarily to predict the physical properties of pure compounds. The problem starts when we apply these equations to mixtures using mixing rules to obtain EOS parameters. Empirical parameters called binary interaction (b.i.) coefficients are employed in some mixing rules to count for the interaction between pairs of different molecules. while these b.i. coefficients may improve the performance of an equation for some multicomponent system, there is no guarantee that the same set of b.i. coefficients will be good for other systems. Better results may be obtained without using any b.i. coefficients at some conditions. In our search for an equation of state for better modelling of gas condensates we gave emphasize to find one which does not use any b.i. coefficients for hydrocarbon mixtures.
P. 535
