Return to search

Laboratory and modelling studies on the effects of injection gas composition on CO₂-rich flooding in Cooper Basin, South Australia.

This Ph.D. research project targets Cooper Basin oil reservoirs of very low permeability (approximately 1mD) where injectivities required for water flooding are not achievable. However, the use of injection gases such as CO₂ would not have injectivity problems. CO₂ is abundant in the region and available for EOR use. CO₂ was compared to other CO₂-rich injection gases with a hydrocarbon content including pentane plus components. While the effect of hydrocarbon components up to butane have been investigated in the past, the effect of n-pentane has on impure CO₂ gas streams has not. One particular field of the Cooper Basin was investigated in detail (Field A). However, since similar reservoir and fluid characteristics of Field A are common to the region it is expected that the data measured and developed has applications to many other oil reservoirs of the region and similar reservoirs elsewhere. The aim of this Ph.D. project is to determine the applicability of CO₂ as an injection gas for Enhanced Oil Recovery (EOR) in the Cooper Basin oil reservoirs and to compare CO₂ with other possible CO₂-rich injection gases. The summarised goals of this research are to: • Determine the compatibility of Field A reservoir fluid with CO₂ as an injection gas. • Compare CO₂ to other injection gas options for Field A. • Development of a correlation to predict the effect of nC₅ on MMP for a CO₂- rich injection gas stream. These goals were achieved through the following work: • Extensive experimental studies of the reservoir properties and the effects of interaction between CO₂-rich injection gas streams and Field A reservoir fluid measuring properties related to: • Miscibility of the injection gas with Field A reservoir fluid • Solubility and swelling properties of the injection gas with Field A reservoir fluid • Change in viscosity-pressure relationship of Field A reservoir fluid due to addition of injection gas • A reservoir condition core flood experiment • Compositional simulation of the reservoir condition core flood to compare expected recoveries from different injection gases • Development of a set of Minimum Miscibility Pressure (MMP) measurements targeted at correlating the effect of nC₅ on CO₂ MMP. The key findings of this research are as follows: • Miscibility is achievable at practical pressures for Field A and similar reservoir fluids with pure CO₂ or CO₂-rich injection gases. • For Field A reservoir fluid, viscosity of the remaining flashed liquid will increase at pressures below ~2500psi due to mixing the reservoir fluid with a CO₂-rich injection gas stream. • Comparison of injection gases showed that methane rich gases are miscible with Field A so long as a significant quantity of C₃+ components is also present in the gas stream. • There is a defined trend for effect of nC₅ on MMP of impure CO₂. This trend was correlated with an error of less than 4%. • Even though oil composition is taken into account with the base gas MMP, it still affects the trend for effect of nC₅ on MMP of a CO₂-rich gas stream. • An oil characterisation factor was developed to account for this effect, significantly improving the results, reducing the error of the correlation to only 1.6%. The significance of these findings is as follows: • An injection pressure above ~3000psi should be targeted. At these pressures miscibility is achieved and the viscosity of the reservoir fluid injection gas mix is reduced. • CO₂ should be compared to gases such as Tim Gas should after considering the cost of compression, pipeline costs and distance from source to destination will need to be considered. • The addition of nC₅ will reduce the MMP and increase the recovery factor, however the cost of the nC₅ used would be more than the value of increased oil recovered. • The developed correlation for the effect of nC₅ on impure CO₂ MMP can be used broadly within the limits of the correlation. • Further research using more oils is necessary to validate the developed oil characterisation factor and if successful, using the same or similar method used to improve other correlations. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1369016 / Thesis (Ph.D.) -- University of Adelaide, Australian School of Petroleum, 2009.

Identiferoai:union.ndltd.org:ADTP/291330
Date January 2009
CreatorsBon, Johannes
Source SetsAustraliasian Digital Theses Program
Detected LanguageEnglish

Page generated in 0.0012 seconds