Academia.eduAcademia.edu

Fluid model

2019

The information, opinions and views expressed in this document do not necessarily represent those of The University of Queensland, the Australian Government or Coal 21. Researchers within or working with the UQ-SDAAP are bound by the same policies and procedures as other researchers within The University of Queensland, which are designed to ensure the integrity of research. The Australian Code for the Responsible Conduct of Research outlines expectations and responsibilities of researchers to further ensure independent and rigorous investigations.

Figures

Figure 1 CO2

Figure 1

Executive summary

When carbon dioxide (CO2) is injected via an injection well and into an aquifer, it goes through different physical processes. From top to bottom of the well, CO2 is heated due to friction, compression, and conduction from the surrounding environment. The pressure rises with depth reaching its supercritical state, at which its density increases by some orders of magnitude.

As injected CO2 flows from the wellbore into the reservoir, it may be heated again and increase in volume if it is cooler than the reservoir. Given its higher compressibility, it is further compressed when displaced by subsequent injected CO2 volumes. At the same time, the injected CO2 is cooling the reservoir with this effect being greatest near the injection well. These transient variations in CO2 density may result in some vertical migration due to buoyancy and may generate thermal convective currents in the reservoir.

At the same time, small portions of the injected CO2 can dissolve in formation water (up to 1.8 moles per kilogram of water or 8% CO2-to-water mass ratio, for up 50 000 kPa) and increase the water density (Appelo, Parkhurst & Post 2014;Akinfiev & Diamond 2010;Duan et al. 2006). The resulting convective currents enhance the mixing of fluids and increase the dissolution.

Fluid model

To calculate the properties and interactions of water and CO2, we have used most of the default models in GEM, but also added some constraints related to the specific reservoir conditions.

Components

We included CH4 and CO2 as components and chose the Peng-Robinson equation-of-state to model their fugacities and volumes, with all input parameters set to default (e.g. specific gravity, critical temperature, acentric factor, etc.). Including CH4 is a numerical condition to control the initial content of CO2 in the reservoir, since GEM requires the sum of mole fractions of gaseous component to be equal to unity. In this way, we could initialise a reservoir with negligible CO2 content by setting 0.999 and 0.001 to CH4 and CO2 mole fractions, respectively. Moreover, we modelled the viscosities using the correlations of Jossi, Stiel, and Thodos, with default coefficients.

Water properties

We have set water compressibility to the default value (4.35E-7 1/kPa at 101.3 kPa), while density and viscosity is calculated via Rowe-Chou and Kestin correlations, respectively. We set water salinity at 3000 ppm of NaCl as Moonie produced water varies between 1000 and 5000 ppm. Water is not allowed to vaporise.

CO2 solubility

We modelled CO2 solubility in water using Henry's Law with Henry's constant determined by the Harvey correlation, making the solubility a function of pressure, temperature and salinity. The CO2 solubility in water is calculated in terms of molality (moles of dissolved gas per kilogram of water)

Thermal option

We enabled heat exchange calculations between fluid components and reservoir rock. Enthalpies were calculated using a polynomial function of temperature for each fluid component and rock thermal properties were set to default values.

10 years of Injection Isothermal (70 °C) 50 °C 35 °C Cum CO2 INJ (Mt)

Cum CO2 INJ (Mm 3 )

CRICOS Provider Number 00025B