Predicting the Properties of Sour Gases and Condensates: Equations of State and Empirical Correlations

Abstract

Abstract Compressibility, destiny and viscosity of natural gases are necessary in most petroleum engineering calculations. Some of these calculations are gas metering, gas compression, design of processing units, and design of pipeline and surface facilities. Properties of natural gases are also important in calculation of gas flow rate through reservoir rock, material balance calculations and evaluation of gas reserves. Usually the gas properties are measured in laboratory. Occasionally, experimental data become unavailable and estimated from equations of state or empirical correlations. This paper presents the results of using various equations of state, corresponding state methods, and correlations to predict the volumetric and transport properties of sour gases and gas condensates. Capabilities of PR-EOS, SRK-EOS, and PT-EOS to predict gas compressibility and density of 2100 gas samples under various schemes of binary interaction number are thoroughly investigated. This study also reports a comparison between modified PR-EOS and other methods to estimate the viscosity of highly sour gases and rich gas condensates. Introduction Natural gas compressibility, density, and viscosity are important properties in the calculations of gas flow through reservoir rocks, material balance calculations, and design of pipelines and production facilities. In the past three decades a number of natural gases and gas condensates fields have been discovered around the world. The major impurities of these natural gases and condensates sources consist of hydrogen sulfide and carbon dioxide. Several methods are now available in literatures for the calculation of natural gas properties. These methods can be classified into three groups1. The first group uses gas composition or gas gravity to calculate pseudo-critical properties of gases and predict gas properties from empirical correlations. In this group, often gas density is used to predict viscosity. Hence prediction of viscosity is dependent on the choice of the method of estimating the density. The second group uses gas composition to estimate gas properties via the method of corresponding states. The third category, the most recent ones, is based on equations of state (EOS) approach. The last category has the advantage of using single equation to calculate k-values, compressibility, density, and viscosity2-4. It also secures stable convergence in the vicinity of the critical point. In EOS-based viscosity models the density calculation is not required for viscosity. Li and Guo5 studied the accuracy of Peng-Robison EOS to predict phase equilibria of sour gases. Because PR-EOS was not accurate, they modified the original PR-EOS by introducing 33 constant. However, this modification makes the equation not convenient for engineering calculations. Mohsen-Nia et al6 introduced a two constant EOS, based on theoretical background of statistical mechanics, designed specially to predict properties of sours natural gases. The equation has several constants (a,ß) for each of the pure components forming the gas mixture. Mohsen- Nia et al. did not explain how to calculate the constants for the plus fraction. They tested their equation for several binary systems and light gases without accounting for the effect of binary interaction numbers (BIN). Huron et al.7 and Evelein and Moore8 used SRK-EOS to study the hydrocarbon system containing hydrogen sulfide and carbon dioxide. They reported phase equilibria calculations but did not report thermodynamic and transport property calculations. Consequently, most of these methods have limited use specially when dealing with sour gases and gas condensates7. In this paper we report the effect of incorporating the BIN on the accuracy of EOS(s) prediction of the properties of sour gases and gas condensate systems. The capabilities of several EOS(s) and several correlations as well as corresponding state methods are thoroughly investigated.