(104c) The Interactions of CO2/Brine/Rock Under CO2 Sequestration Conditions

Authors: 
Howard, B. H., U.S. DOE/NETL
Crandall, D., NETL
Dilmore, R., U.S. Department of Energy
Zhang, L., DOE/NETL

The Intergovernmental Panel on Climate Change has recently reported that injection of CO2 into confined geological formations, given their potentially massive carbon storage capacity and widespread geographic distribution, represents one of the most promising options for mitigating anthropogenically-induced global climate change.  The lower Tuscaloosa Formation in the southeast region of the U.S is recognized as a promising candidate host reservoir for carbon sequestration. The lower Tuscaloosa formation is a deep saline aquifer and is a primary reservoir target for large scale carbon dioxide injection tests due to its proximity to CO2 sources, favorable depth, thickness, permeability, porosity, and the presence of an overlying low permeability shale formation as a seal.  The Southeast Regional Carbon Sequestration Partnership has selected the lower Tuscaloosa Formation as one of its major demonstration projects. While the lower Tuscaloosa Sandstone has ideal reservoir characteristics in some areas, there are significant variations in porosity, permeability, and mineralogy.

Numerous studies have been conducted to investigate changes in host rock properties when exposed to CO2.  Luquot and Gouze (2009) conducted CO2-enriched fluid flow through a carbonate core sample under in situ sequestration conditions (T = 100 °C and P = 12 MPa).  Under these conditions, a decrease in permeability and porosity is seen that is linked to precipitation of Mg-rich calcite.    Gouze and Luquot (2011) also conducted an X-ray microtomography study of porosity, permeability and reactive surface changes during the dissolution of pure calcite in a brine-CO2 mixture.  They showed that the increase in permeability is due to the decrease of tortuosity for homogeneous dissolution, whereas it is due to the combination of tortuosity decrease and hydraulic radius increase for heterogeneous dissolution.  Jove Colon et al., (2004) studied the permeability, porosity and reactive surface area evolution during dissolution of non-fractured, clay-free Fontainebleau sandstone cores using a flow-through percolation reactor. The experiments were performed at 80°C with 0.1 M NaOH solution.  The permeability was found to vary depending on where the dissolution occurred: in the pore void or the pore throat.  Hence, the injection of CO2 into a deep saline aquifer will affect porosity since it can cause both mineral dissolution and precipitation in the formation.  To predict the most likely changes to the reservoir porosity and permeability during sequestration, it is critical to understand the behavior of the minerals that line pore spaces when they are in contact with the CO2 saturated brine.

An experimental study was conducted to assess the potential impacts of injected CO2. Core samples from the lower Tuscaloosa Formation were placed in CO2 saturated brines at sequestration temperature of 85 °C and pressure of 23.8 MPa in the presence of synthetic brine for a period of six months in 1.3 liter high-pressure vessels. Core samples from Massive Sandstone, Marine Shale, and Mudstone from the Middle Tuscaloosa formation were used in this study. Computed tomography (CT), micro CT, XRD, SEM, petrography, porosity, and permeability analyses were conducted prior to the six month exposure.  Upon completion of the experiment, these analyses were repeated.

 

Preliminary permeability measurements obtained from the core sample showed an obvious change after exposure to CO2-saturated brine for six months. This observation suggests that mineral dissolution and mineral precipitation could occur in the host deposit altering its characteristics for CO2 storage over time.  Examination of the same locations on the surface of samples prior to and following exposure suggested that some mineral dissolution ad precipitation took place, for example, Ba-containing Celestine precipitation was observed on the mudstone sample.