Semianalytical Solution for CO2 Leakage through an Abandoned Well

详细信息    查看全文

• 作者：Jan Martin Nordbotten ; Michael A. Celia ; Stefan Bachu ; and Helge K. Dahle
• 刊名：Environmental Science & Technology
• 出版年：2005
• 出版时间：January 15, 2005
• 年：2005
• 卷：39
• 期：2
• 页码：602 - 611
• 全文大小：380K
• 年卷期：v.39,no.2(January 15, 2005)
• ISSN：1520-5851

文摘

Capture and subsequent injection of carbon dioxide intodeep geological formations is being considered as a meansto reduce anthropogenic emissions of CO₂ to theatmosphere. If such a strategy is to be successful, theinjected CO₂ must remain within the injection formation forlong periods of time, at least several hundred years.Because mature continental sedimentary basins have acentury-long history of oil and gas exploration and production,they are characterized by large numbers of existing oiland gas wells. For example, more than 1 million such wellshave been drilled in the state of Texas in the UnitedStates. These existing wells represent potential leakagepathways for injected CO₂. To analyze leakage potential,modeling tools are needed that predict leakage rates andpatterns in systems with injection and potentially leakywells. A new semianalytical solution framework allows simpleand efficient prediction of leakage rates for the case ofinjection of supercritical CO₂ into a brine-saturated deepaquifer. The solution predicts the extent of the injected CO₂plume, provides leakage rates through an abandonedwell located at an arbitrary distance from the injectionwell, and estimates the CO₂ plume extent in the overlyingaquifer into which the fluid leaks. Comparison to resultsfrom a numerical multiphase flow simulator show excellentagreement. Example calculations show the importanceof outer boundary conditions, the influence of both densityand viscosity contrasts in the resulting solutions, andthe potential importance of local upconing around the leakywell. While several important limiting assumptions arerequired, the new semianalytical solution provides a simpleand efficient procedure for estimation of CO₂ leakage forproblems involving one injection well, one leaky well, andmultiple aquifers separated by impermeable aquitards.