Summary
In terms of experimentation, acoustic velocity can be measured with a high degree of accuracy. Several thermodynamic properties related to acoustic velocity such as density, isothermal compressibility, and heat capacity can be extracted from measured data. In this study, technical improvements are implemented in an effort to develop a technique for fast and reliable determination of fluid properties on the basis of acoustic velocity measurements over an expanded range of pressures. The potential use of this device as a quality-control tool in typical pressure/volume/temperature (PVT) measurements is demonstrated.
Baseline measurements matched to published literature verify the suitability of the device. Results of tests on three recombined oil samples containing dissolved gas, with prescribed gas/oil ratios (GORs), and one bitumen sample are presented. A sharp change in the acoustic velocity trend near the gas/liquid-saturation point is evidence of gas evolution during depressurization. Strong attenuation complicates measurement of acoustic velocity on the heavy fluids used in this study. Blending bitumen with a midrange-molecular-weight hydrocarbon mixture enables estimation of the undiluted-fluid acoustic velocity by extrapolation.
By use of the measured acoustic velocity data available, a methodology is developed to estimate and quality check measured isothermal compressibility (κT) values. This is especially important for low-compressibility systems. Heat-capacity data for simple alkanes (CH4 to n-C10) and toluene helps to define a reasonable range of heat-capacity ratio (γ) expected for typical reservoir fluids. For the typical values of acoustic velocity encountered in the pressure and temperature range of interest, the isothermal compressibility can be calculated and/or quality checked by use of estimated values of γ. In addition, by use of various data sets and by performing graphical error analysis, we have shown the reasons that the methodology works. Available data for n-decane and n-hexadecane along with measured data for a live oil and numerical work on calibrated data sets in this study are used to develop the methodology.
