Computing Sciences masthead Berkeley Lab Computing Sciences Berkeley Lab logo
Share/Bookmark

Exascale for Energy


Carbon Capture and Sequestration

Putting the brakes on CO2 emissions

Carbon capture and sequestration—injecting carbon dioxide (CO2) into the Earth’s subsurface instead of releasing it into the atmosphere—is one of the most promising ways for reducing the buildup of greenhouse gases in the atmosphere.  In fact, even under the most optimistic scenarios for energy efficiency gains and the greater use of low- or no-carbon fuels, sequestration will likely be essential if the world is to stabilize atmospheric concentrations of greenhouse gases at acceptable levels.


Three computational research projects are helping answer questions about the behavior of CO2 in geologic formations and the most economical ways to capture CO2 at its sources.

CO2 sequestration in geologic formations includes oil and gas reservoirs, unmineable coal seams, and deep saline reservoirs. These are structures that have stored crude oil, natural gas, brine, and CO2 over millions of years. Many power plants and other large emitters of CO2 are located near geologic formations that are amenable to CO2 storage. In many cases, injection of CO2 into a geologic formation can enhance the recovery of hydrocarbons, providing value-added byproducts that can offset the cost of CO2 capture and sequestration.

While pilot projects have demonstrated the feasibility of carbon sequestration, there are unanswered questions about the behavior of CO2 in geologic formations and the most economical ways to capture CO2 at its sources. Three research projects in Berkeley Lab’s Computational Research Division (CRD) are helping to find the answers to those questions. All of the projects involve collaborations with one or more of DOE’s Energy Frontier Research Centers (EFRCs).

When CO2 is pumped into geologic reservoirs for long-term storage, acidic fluids resulting from CO2 reacting with groundwater and brine may dissolve some of the minerals in the reservoirs and deposit them downstream as carbonates, changing the pore geometry and flow patterns of the geologic formation. Modeling this process mathematically is the goal of a project called “Advanced Simulation of Subsurface Flow and Transport at the Pore Scale,” led by Phillip Colella of CRD.

This project, a collaboration between the SciDAC Applied Partial Differential Equations Center for Enabling Technology (APDEC) and the Energy Frontier Research Center for Nanoscale Control of Geologic CO2 (NCGC), will develop the algorithmic and software infrastructure tools to enable NCGC’s goal of modeling molecular- to pore-scale processes in geologic systems. Specifically, the team will develop algorithms and software to model multiphase, reacting flow of CO2 and water in a complex heterogeneous medium, with modification of microscale pore structures by mineral dissolution and precipitation. The Chombo numerical algorithm software package supported by APDEC will be extended so it can be applied to this class of problems.

pore structures in zeolites
Figure 15. A sampling of pore structures in zeolites, a type of microporous mineral. A research project at Berkeley Lab will automate the computational screening of 2.5 million theoretically possible zeolite structures to search for materials that can economically capture CO2 at its sources.

Another collaboration, between NCGC and the SciDAC Visualization and Analytics Center for Enabling Technology (VACET), led by E. Wes Bethel of CRD, will give NCGC researchers the necessary tools to visualize and analyze their data effectively and improve their understanding of processes governing carbon sequestration. The project “Visualization and Analysis for Nanoscale Control of Geologic CO2” intends to bridge the gap between experimental data and numerical simulations by developing image processing and analysis tools to automate measurements in both experimental and simulated data, and by developing geometric analysis techniques to extract relevant features from numerical simulation data that can be compared to experimental data. The tools developed in this project will accelerate data processing and provide new capabilities for analysis of simulations, thus allowing NCGC researchers to run more experiments and effectively target new experiments.

The cost of capture is one of the main bottlenecks for large-scale carbon capture and sequestration. For example, the conventional technology for capturing CO2 from the effluent stream of a power plant may require as much as 25% of the electricity being produced. The EFRC for Gas Separations Relevant to Clean Energy Technologies aims to tackle this problem by developing novel porous materials, such as zeolites or metal organic frameworks, that can capture CO2 economically. But finding the ideal pore topologies from the 2.5 million theoretically possible zeolite structures (Figure 15) requires new computational tools for automated structural analysis.

In the project “Accelerating Discovery of New Materials for Energy-Related Gas Separations through PDE-Based Mathematical and Geometrical Algorithms and Advanced Visualization Tools,” also led by Bethel, the SciDAC VACET center will collaborate with this ERFC to develop state-of-the-art algorithms for analyzing and screening chemical systems, using a combination of advanced partial differential equation (PDE) algorithms to detect and probe geometric structures, and visualization techniques to track the motion of chemical probes through complex structures. Moving these algorithms to high performance computing platforms and deploying advanced data analysis tools will allow the researchers to automatically screen millions of pore structures without human intervention. The most promising structures can then be synthesized and tested.

 


<< Previous page