Chen, Ganglin (ExxonMobil Upstream Research Co.) | Chu, Dez (ExxonMobil Upstream Research Co.) | Zhang, Jie (ExxonMobil Upstream Research Co.) | Xu, Shiyu (ExxonMobil Upstream Research Co.) | Payne, Michael A. (ExxonMobil Upstream Research Co.) | Adam, Ludmila (Colorado School of Mines) | Soroka, William L. (ADCO)
Bächle, Gregor (University of Miami) | Eberli, Gregor (University of Miami) | Madadi, Mahyar (Australian National University) | Sok, Rob (Australian National University) | Knackstedt, Mark A. (Australian National University) | Arns, Christoph (Australian National University) | Latham, Shane (Australian National University) | Sheppard, Adrian P. (Australian National University)
Lebedev, Maxim (Curtin University of Technology) | Gurevich, Boris (Curtin University of Technology) | Toms, Juliana (Curtin University of Technology) | Clennel, Ben (CSIRO-petroleum) | Pervukhina, Marina (CSIRO-petroleum) | Mueller, Tobias (University of Karlsruhe)
Tutuncu, Azra (Shell Exploration and Production Company) | Rojas, Maria Alejandra (University of Houston) | Castagna, John (University of Houston) | Krishnamoorti, Ramanan (University of Houston) | Han, De Hua (University of Houston)
Visco elastic measurements of four heavy and extra-heavy oil samples were carried out to analyze the dependence of complex viscosity, loss and storage modulus with temperature and frequency. The dynamic rheological tests showed a shear thinning phenomenon typical of non- Newtonian fluids, highly pronounced for seismic and sonic frequencies and for temperatures below 50°C. The Power- Law method that explains the shear thinning behavior was modified to incorporate the liquid crystal theory and the viscosity dependence on temperature based on the concept of activation energy. An expression was derived to predict complex viscosity based on frequency and temperature changes.
The need to characterize fluid flow properties of unconventional reservoirs, such as heavy and extra-heavy oils, has increased significantly in the last few years. Heavy oil properties are particularly dependent on frequency and temperature changes. According to its rheological properties, it can be considered a Non-Newtonian viscoelastic fluid, which means shear stress and shear strain rate are not linearly correlated. Since such materials have an elastic component, they are able to support shearing. Previous ultrasonic measurements have shown that shear wave velocity dispersion might be significant for heavy oils at in-situ conditions. Velocity dispersion (for P- and S waves) seems to be negligible for heavy oils in the liquid and glass (elastic) phases (Han et al., 2007).
There have been numerous studies in the past on crude oils concerning the shear-thinning (complex viscosity decrease as shear strain rate increases) behavior obeying the Power- Law model at reservoir temperatures (Wang et al., 2006). Above a certain temperature, sometimes called the liquid point temperature (Han et al., 2006), heavy oil becomes Newtonian and viscosity becomes independent of the shear strain rate (or frequency).
This paper aims to model the shear thinning behavior of heavy oil at a given temperature as frequency increases by using the concept of activation energy of the Arrhenius equation and liquid crystal theory. Laboratory measurements of the viscoelastic properties of four heavy and extra-heavy oil samples are used to improve the model.
Theory and experiment procedure
For non-Newtonian viscoelastic materials, the complex shear modulus (G*) is given by the in-phase elastic component or storage modulus (G'') and the out of phase viscous component or loss modulus (G"). G'' and G" represent the ability for a material to store energy elastically and to dissipate energy respectively.
Viscosity versus Temperature
The complex viscosity trend with temperature for Newtonian and viscoelastic fluids indicates that ?* decreases significantly as temperature increases at low temperatures, while for high temperatures ?* decreases slowly with temperature at high temperatures. This is clearly shown by the DeGuetto empirical model (Figure 1), which predicts similar ?*-T trend but with considerably lower viscosities.
Measured data reflects three stages. In the first stage viscosity decreases rapidly by 2 orders of magnitude for temperatures below 30°C. For temperatures between 40 and 50°C, the viscosities decrease more slowly. At higher temperatures, ?* decreases faster than stage 2 but lower than stage 1 (Figure 1). Results for Sample 1 were repeated using two different rheometers giving similar response.