Liquid-Density Correlation of Hydrocarbon Systems

Abstract

Published in Petroleum Transactions, AIME, Volume 219, 1960, pages 386–389. Abstract The Standing-Katz method for predicting liquid densities of reservoir fluids has been tested using experimental data of 154 bottom-hole or recombined reservoir fluid samples. New pressure- and temperature-correction curves for the Standing-Katz type correction are presented which improve the accuracy of prediction. Introduction The present study was undertaken to investigate the known methods of determining liquid densities and to select one of them which might yield a better correlation if additional data were employed. An appropriate correlation would incorporate the following characteristics:ease of handling in the computations,results within engineering accuracy anddata easily accessible. After careful investigation of the different methods, the Standing-Katz correlation appeared to have these characteristics. The Standing-Katz correlation for liquid densities of hydrocarbon systems is based upon limited data. The original work, which employs apparent densities for methane and ethane, was based on data of 15 saturated crude oils in equilibrium with natural gas. Therefore, if test of the correlation using additional experimental data would serve either to validate the Standing-Katz correlation or to form a basis for corrections if such were necessary. Procedure and Graphical Solution The data of 154 bottom-hole or recombined reservoir fluid samples were employed to calculate the densities at 14.7 psia and 60 degrees F by the method described in Table 1. Volume contributions for methane and ethane were assigned from the pseudo liquid density plot (Fig. 1), but volume contributions for N2, CO2 and H2S were not considered. The densities computed at reference conditions of 14.7 psia and 60 degrees F were then elevated to their respective temperatures and pressures by the use of the correction curves proposed by Standing. The percentage differences between the experimental values and the calculated values were determined: these samples were grouped into temperature, pressure and density ranges to determine a possible trend. It appeared that, at temperatures above 160 degrees F, the calculated density values were consistently larger than the experimental density values; however, no obvious trend was found for the density and pressure ranges.