In the OSTI Collections: Carbon Sequestration
Dr. William N. Watson, Physicist
DOE Office of Scientific and Technical Information
“… the Michigan Geological Repository for Research and Education (MGRRE) … continues the original Michigan Basin Core Research Laboratory's mission of geological research and technology transfer about Michigan subsurface geology, efficient recovery of fossil fuels and carbon dioxide sequestration. It also includes geological research and education about water resources and environmental topics. … There are also extensive web-based databases for information dissemination. The facility functions as a primary location for research and technology transfer related to petroleum geology, CO2 sequestration, Great Lakes bluff erosion and water resources. ...”
—from the home page of the Michigan Geological Repository for Research and Education
The sun supplies nearly all the power that drives the earth’s weather and supports life on it. Some of the energy we receive is reflected back out into space immediately, but the rest is absorbed and produces one or more other effects before eventually being radiated into space again.
Two atmospheric gases that play important roles for life on earth, water vapor and carbon dioxide, reflect some of this reradiated energy back toward the earth’s surface, slowing its eventual return to space. With energy staying near the earth longer before leaving, more energy is here at any given time, making things warmer than they’d be otherwise. Other things being equal, increasing the total amount of carbon dioxide in the atmosphere leads to even more energy being retained for still longer, causing greater average warmth.
Numerous measures are being studied and taken to halt and reverse the increase of atmospheric carbon dioxide, including ways to remove some of the carbon dioxide from the atmosphere and put it someplace else where it won’t slow the return of energy into space. Recent reports available through OSTI’s SciTech Connect provide information about examples of these carbon sequestration efforts, which range from understanding chemical reactions involved in carbon-dioxide removal to informing carbon-reservoir design by computer simulation of entire reservoirs.
One place to sequester CO2 is in underground reservoirs. The effectiveness of putting it there depends on exactly how the carbon dioxide interacts with the reservoirs’ minerals and water. Three recent studies to understand those interactions are described in reports by researchers at Sandia National Laboratories, Stanford University, and a collaboration from the George Washington University and the University of Wisconsin-Madison.
The Sandia report’s focus is indicated by its title, “Fundamental study of CO2-H2O-mineral interactions for carbon sequestration, with emphasis on the nature of the supercritical fluid-mineral interface”[SciTech Connect]. How permeable a reservoir is to CO2 may depend on how saturated with water the reservoir is. The report describes how a theoretical model of water absorption/condensation was confirmed in its major conclusions by experiments. When the CO2 is at a high-enough temperature and pressure to be in its supercritical state (i.e., when its liquid and gaseous states merge), the water film that forms in it is reactive with common rock-forming minerals to a degree that depends on the water-film thickness. At one level of water activity, the greatest extent of reaction in supercritical CO2 was found to occur in places like step edges and surface pits, where capillary condensation[Wikipedia] thickened the water films. According to the report, “This suggests that dissolution/precipitation reactions may occur preferentially in small pores and pore throats, where it may have a disproportionately large effect on rock hydrologic properties.” The report also presents a theoretical model that allows assessment of the ways pore size and structural heterogeneity affect capillary trapping efficiency, which suggest approaches to optimizing trapping capacity.
The report by George Washington University and U. Wisconsin-Madison researchers, “Interface Induced Carbonate Mineralization: A Fundamental Geochemical Process Relevant to Carbon Sequestration”[SciTech Connect], discusses reactions that might sequester carbon dioxide, namely the formation of magnesite (MgCO3) and dolomite (MgxCa(1-x)CO3 for x=0.5). However, these formation reactions present a problem: minerals high in magnesium carbonate can’t be formed at ambient low-temperature and low-pressure conditions. Hydration is known to be one hindrance to this process, and an important one since water is expected in natural reservoirs. But to see what else besides hydration might hinder the process, the authors studied the formation of MgCO3 and MgxCa(1-x)CO3 in waterless solutions. The authors’ experimental data suggested to them that the various atoms constituting these minerals can’t easily get into the proper alignment to form them under ordinary ambient conditions. “Unless the Mg-O octahedra are efficiently compressed so that the CO3 group can have more freedom in motion,” they note, “… the energy barrier for forming an ordered Mg-CO3 arrangement may be too high to be overcome at low temperature/pressure conditions…. On the other hand, the measured higher distribution coefficients of Mg between magnesian calcites formed in the absence and presence of water give us a first direct proof to support and quantify the cation hydration effect. These findings may expand our current understanding of the Mg(-Ca)-CO3 system and provide important insight into dolomite/magnesite formation as well as the processes involved in biomineralization and mineral carbonation.”
Figure 1. [From “Interface Induced Carbonate Mineralization: A Fundamental Geochemical Process Relevant to Carbon Sequestration”[SciTech Connect], p. 9.]
Injecting and storing carbon dioxide in rocks will affect the rocks’ elastic properties in ways that may limit how useful the process is. The Stanford University report “Rock Physics of Geologic Carbon Sequestration/Storage”[SciTech Connect] presents a theory of these effects, specifically addressing how the elastic properties of a CO2 reservoir’s rocks are affected by (a) CO2 saturation and pore pressure combined, (b) CO2 saturation and rock-fabric alteration combined, and (c) CO2 saturation and the CO2’s distribution after injection, and how (d) the rocks’ attenuation of seismic waves[Wikipedia] varies with CO2 saturation and subsequent distribution. Results of the Stanford research include a new method of applying the theory to well data and a determination from the rocks’ elastic changes of how the rocks will alter seismic waves.
Figure 2. Scanning electron microscope images showing changes to a portion of rock after CO2 injection. Left: before injection; right: after. [From “Rock Physics of Geologic Carbon Sequestration/Storage”[SciTech Connect], p. 47.]
The abstract of “Rock Physics of Geologic Carbon Sequestration/Storage” ends by stating, “As planned, three graduate students participated in this work and, as a result, received scientific and technical training required should they choose to work in the area of monitoring and quantifying CO2 sequestration.” Training people in the field of carbon sequestration was a significant objective in many of the recently reported projects—in some cases even the chief objective. As “Recovery Act: Multi-Objective Optimization Approaches for the Design of Carbon Geological Sequestration Systems”[SciTech Connect] from Colorado State University says about the design project it describes, “The main objective of this project is to provide training opportunities for two graduate students in order to improve the human capital and skills required for implementing and deploying carbon capture and sequestration (CCS) technologies.” Projects like this one develop students’ skills through their production of something useful. The Colorado state students’ product is to be an integrated simulation-optimization framework to support the design of systems to capture and sequester carbon that will maximize carbon storage while minimizing the systems’ total costs as well as the risk of CO2 leaking upward from injected formations. Figure 3 shows examples of the kind of system to be mathematically modeled and optimized, while Figure 4 is a graphic representation of how the cost and risk minimization and the carbon-storage maximization balance each other out.
Figure 3. Top: Vertical cross section of a typical reef formation situated in the Michigan geological basin along with reef facies distribution chart. Note that, to show the formation’s layers more distinctly, this diagram’s vertical scale is 10 times the size of its horizontal scale. Bottom: Schematic illustration of the Michigan Technological University’s Test Site in the northern reef trend of Michigan. [From “Recovery Act: Multi-Objective Optimization Approaches for the Design of Carbon Geological Sequestration Systems”[SciTech Connect], p. 51; left portion of figure on that page is adapted from “The Belle River Mills Gas Field: Productive Niagaran Reefs Encased by Sabkha Deposits, Michigan Basin” by D. Gill, Michigan Basin Geological Society, Special Papers No.2 (1977).]
Figure 4. A surface that represents the relation between the cost, CO2-leakage risk, and amount of CO2 sequestered for a carbon-dioxide sequestration system at a particular site. Here, the CO2-leakage risk, a quantity to minimize, is represented by the probability that the system’s capacity won’t be exceeded (“Non-Exceed Prob.”), a quantity to maximize. Project cost increases with both non-exceedance probability and mass of CO2 sequestered. Project cost sharply increases with increasing non-exceedance probability above 70%. This demonstrates that the magnitude of uncertainty associated with passive well permeability results in a high degree of project cost uncertainty for this site. In addition, two plateaus are observed on the top of the surface showing regions where the mass of CO2 sequestered may be greatly increased for relatively small increases in project cost. [From “Recovery Act: Multi-Objective Optimization Approaches for the Design of Carbon Geological Sequestration Systems”[SciTech Connect], p. 26.]
If there is a single best way to accomplish something that people are just beginning to have some experience with, that way is usually not known at the time, so it pays to try several different approaches. A Colorado School of Mines project, reported in “Training and Research on Probabilistic Hydro-Thermo-Mechanical Modeling of Carbon Dioxide Geological Sequestration in Fractured Porous Rocks”[SciTech Connect], has quite similar overall objectives of sequestration simulation and risk assessment as the Colorado State University project does. But since the School of Mines project involves different people, their model will be informed by their similar but nonidentical relevant knowledge and experiences. The design from this project, combined with those of the Colorado State project and other similar ones, should eventually result in a more accurate and useful model than any one effort alone would have done. The other result of this project, training of the graduate students who develop the model, augments the similar results of other projects by increasing even further the number of people who can advance carbon-dioxide sequestration.
A third project, reported in “Recovery Act: Geologic Sequestration Training and Research”[SciTech Connect] by authors affiliated with the University of Alabama at Birmingham, Southern Company, Jefferson State Community College, and the John A. Volpe National Transportation Systems Center, involved a wider range of knowledge and research relevant to carbon-dioxide sequestration. Besides the similar simulation of carbon-dioxide migration in storage reservoirs and seepage through seal layers, this project involved
• establishing a rock-properties measurement lab,
• evaluating caprocks’[Wikipedia] sealing capacity and reservoirs’ porosity, permeability, and storage capacity,
• developing an advanced course on coal combustion and gasification, climate change, and carbon sequestration, which was offered twice during the project period (December 2009 through June 2013) and completed by a total of 49 students.
Students also did independent research on rock properties and reservoir simulation. The project involved four graduate students and one undergraduate student. Two graduate students were awarded doctorates, another proposed research on an advanced porosity-measurement technique and was admitted to the doctoral program, and the other prepared a proposal for research on carbon capture, use, and storage and solid-waste management. The undergraduate student performed measurements on caprock and reservoir rock samples, and received a bachelor’s degree in mechanical engineering.
A fourth report[SciTech Connect] describes the establishment by Western Michigan University investigators of a research and education center named MichCarb which particularly emphasizes technology transfer. Western Michigan University is the home of the Michigan Geological Repository for Research and Education quoted at the beginning of this article. The establishment of MichCarb involved several different things:
• Archiving and maintaining a current reference collection of carbon-sequestration publications
• Developing statewide and site-specific digital research databases for deep geological formations in Michigan that are relevant for carbon-dioxide storage, containment, and potential for enhanced oil recovery
• As components of these databases, producing maps and tables of physical properties
• Compiling all information into a digital atlas
• Mathematically modeling geology and fluid flow, compiling and integrating data for the models, and applying the models to specific predictive uses for carbon-dioxide storage and enhanced oil recovery
• Characterizing Michigan’s oil and gas and saline reservoirs for carbon-dioxide storage potential volume, injectivity, and containment
• Effecting technology transfer to members of industry and government agencies through a website, publications in relevant journals, and workshops
The report also describes MichCarb research focused on the Michigan Basin[Wikipedia] which, the authors concluded, has excellent geological carbon-sequestration potential with substantial, associated incremental oil-production potential. The report quantifies this potential in detail, as indicated by Figure 5.
Figure 5. CO2 storage capacities in Michigan, by county. Left: Estimated storage capacity for Mount Simon Sandstone. Bottom left: Sandstone, mix and tripolitic chert lithology (conventional and unconventional reservoirs) storage capacity assuming 4% efficiency, with composite net porosity contours superimposed. Bottom right: Sandstone and mix lithologies (conventional reservoirs only) storage capacity assuming 4% efficiency, with composite net porosity contours superimposed. [From “Establishing MICHCARB, a geological carbon sequestration research and education center for Michigan, implemented through the Michigan Geological Repository for Research and Education, part of the Department of Geosciences at Western Michigan University”[SciTech Connect], pp. 40, 46.]
These last two educational projects involved modeling specific sites to estimate their potential. Such modeling is also done at national labs. The Los Alamos National Laboratory report “LANL Deliverable to the Big Sky Carbon Sequestration Partnership: Preliminary CO2-PENS model”[SciTech Connect] describes the results of one such project, a preliminary subsurface risk assessment of Toole County, Montana’s Kevin Dome, a large geologic feature whose area is roughly that of London. The system modeling is done with CO2-PENS and related software. The report lists 138 features, events, and processes relevant to risk assessment for Kevin Dome, notes the four of them that were accounted for in the preliminary study, and describes plans to carry the project forward as additional relevant data becomes available.
The last two reports deal with a quite different way to remove CO2 from the atmosphere, using a particular stage in the natural carbon cycle—carbon’s routing through plants, particularly in forests. Two recent reports, one each from the State University of New York at Albany and the University of Missouri, deal with understanding different aspects of this process.
Since plants photosynthesize sugars using energy from light and carbon atoms taken from carbon dioxide, the quality of sunlight shining on a forest is an indicator of how quickly the forest’s plants are removing carbon dioxide from the air. The quality of solar radiation depends in a complex way on local cloud cover. Clouds reduce the light level (which, other things being equal, would reduce photosynthesis), but they also make the light more diffuse (which would increase photosynthesis), as well as producing lighter and darker periods and affecting which frequencies of light reach the forest at all. Determining exactly how cloud cover affects photosynthetically active radiation is the point of the research described in the Albany report “Evaluating the Contribution of Climate Forcing and Forest Dynamics to Accelerating Carbon Sequestration by Forest Ecosystems in the Northeastern U.S.”[SciTech Connect]. Using data for Harvard Forest in Massachusetts and Tapajos National Forest in Brazil, the authors found how the distribution of lighter and darker periods’ durations depended on whether the clouds were low-level or mid-to-high level, and on how much of the sky each level of clouds covered. This quantification is just one step towards determining the role of Northeastern U.S. forests in sequestering carbon; the report lists further steps as tasks for future work.
Figure 6. Albany, NY, 19 October 2000. Upper graph: Celiometer cloud height in meters. Middle graph: Sky camera cloud cover in percent. Lower graph: Diffuse fraction of the photosynthetically active radiation.
Arrows in the graph indicate 1215 local time, when images below were taken (unprocessed image, bottom left; processed image, bottom right) and clouds were predominantly cumulus. [After “Evaluating the Contribution of Climate Forcing and Forest Dynamics to Accelerating Carbon Sequestration by Forest Ecosystems in the Northeastern U.S.”[SciTech Connect], p. 7.]
Relative frequencies of light and dark periods of various durations under conditions of scattered-cloud and broken-cloud prevelance[Wikipedia] at low levels and at mid-to-upper levels, as found using data for Harvard Forest in Massachusetts and Tapajos National Forest in Brazil. [From “Evaluating the Contribution of Climate Forcing and Forest Dynamics to Accelerating Carbon Sequestration by Forest Ecosystems in the Northeastern U.S.”[SciTech Connect], pp. 23-24.]
Finally, the University of Missouri report shows that some of the other relevant variables have been measured in a different forest and used to check mathematical models of its matter and energy fluxes. During the project described by “Regulation of Carbon Sequestration and Water Use in a Ozark Forest: Proposing a New Strategically Located AmeriFlux Tower Site in Missouri”[SciTech Connect], data streams of CO2 fluxes, water vapor fluxes, and numerous weather variables were established and measurements from them were submitted to a data archive. Among the other variables measured were some related to leaf biochemistry and physiology, biomass inventory, and tree dimensions[Wikipedia]. The investigators also improved a previous ecosystem model to fully treat all major biophysical fluxes, soil organic matter pool dynamics, soil hydrology, and heat transfer. Major findings listed in the report relate to carbon uptake in the forest, soil respiration, use of water by trees, responses of different species to drought (which occurred during the project), variables affecting surface energy partitioning, the importance of root distribution to ecosystem fluxes, and long-term vegetation dynamics.
• Okta (unit of cloud coverage)
Reports Available through OSTI’s SciTech Connect
• “Fundamental study of CO2-H2O-mineral interactions for carbon sequestration, with emphasis on the nature of the supercritical fluid-mineral interface” [Metadata and full text available through OSTI’s SciTech Connect]
• “Interface Induced Carbonate Mineralization: A Fundamental Geochemical Process Relevant to Carbon Sequestration” [Metadata and full text available through OSTI’s SciTech Connect]
• “Rock Physics of Geologic Carbon Sequestration/Storage” [Metadata and full text available through OSTI’s SciTech Connect]
• “Recovery Act: Multi-Objective Optimization Approaches for the Design of Carbon Geological Sequestration Systems” [Metadata and full text available through OSTI’s SciTech Connect]
• “Training and Research on Probabilistic Hydro-Thermo-Mechanical Modeling of Carbon Dioxide Geological Sequestration in Fractured Porous Rocks” [Metadata and full text available through OSTI’s SciTech Connect]
• “Recovery Act: Geologic Sequestration Training and Research” [Metadata and full text available through OSTI’s SciTech Connect]
• “Establishing MICHCARB, a geological carbon sequestration research and education center for Michigan, implemented through the Michigan Geological Repository for Research and Education, part of the Department of Geosciences at Western Michigan University” [Metadata and full text available through OSTI’s SciTech Connect]
• “LANL Deliverable to the Big Sky Carbon Sequestration Partnership: Preliminary CO2-PENS model” [Metadata and full text available through OSTI’s SciTech Connect]
• “Evaluating the Contribution of Climate Forcing and Forest Dynamics to Accelerating Carbon Sequestration by Forest Ecosystems in the Northeastern U.S.” [Metadata and full text available through OSTI’s SciTech Connect]
• “Regulation of Carbon Sequestration and Water Use in a Ozark Forest: Proposing a New Strategically Located AmeriFlux Tower Site in Missouri” [Metadata and full text available through OSTI’s SciTech Connect]