By Will Keener
Can we pump carbon dioxide underground and store it there to avoid warming the atmosphere? That is the question researchers at Sandia, teaming with 40 industrial partners and 10 technical partners (including state geologists, oil conservation departments, geological surveys, and the like) will attempt to answer over the next several years.
Following a phase one study, which showed a number of potential storage sites in the Southwest region, work toward phase two demonstrations is now under way, says David Borns, manager of Sandia’s Geotechnical and Engineering Dept. 6113. “Phase two will demonstrate at the field scale that sequestration really is possible with power plant outputs of CO2 injected in several types of geologic environments. In fact, multiple projects are possible in the region,” says David.
The Southwest Regional Partnership for Carbon Sequestration, which includes Arizona, New Mexico, Colorado, Utah, Texas, Oklahoma, Kansas, and Wyoming, has proposed a series of validation tests of the most promising sequestration technologies, including three major geologic tests and two terrestrial pilot tests.
Geologic options include pumping CO2 into: (1) oil reservoirs to increase oil recovery rates, (2) coalbed methane zones where coal absorbs the carbon dioxide and releases methane gas, and (3) aquifers, where the CO2 combines with water stored in pore spaces in the rock. Terrestrial tests will determine if natural photosynthesis activity can be increased to tie up more CO2 from the atmosphere, David says.
DOE, spurred by a presidential goal of reducing carbon emissions by 18 percent, began an evaluation of sequestration two years ago, funding regional studies in seven regions. Three regions were selected for phase two projects. The Southwest regional project is funded at $16 million over four years, with New Mexico Tech acting as coordinator for the work.
“If carbon sequestration proves effective in managing global warming impacts, some of the first options are likely to coincide with existing CO2 transportation infrastructure,” says David. The Southwest region is home to an extensive CO2 pipeline network, transporting more than 30 metric tons of natural CO2 from the central Rockies to the Permian Basin, where it is used for enhanced oil recovery.
“Our phase one study concluded that the ‘lowest hanging fruit’ for sequestration would be to supplant the natural CO2 with power-plant sourced CO2,” says David. The partnership’s proposal includes:
A complex issue
Technically speaking, CO2 sequestration is a complex issue spanning a wide range of scientific, technological, economic, safety, and regulatory issues, says John Lorenz (6116). John and a team of Sandians will conduct detailed studies on the long-term geologic impacts of CO2 on the host reservoir to determine what characterization work will be needed before sequestration can be deployed.
“The overall objective of the project is to better understand CO2 sequestration-related processes and to predict and monitor the migration and ultimate fate of CO2 after it’s injected into a reservoir,” says John. Although saline aquifers, deep coal seams, depleted gas reservoirs and several other potential reservoirs are available, depleted oil reservoirs make an attractive option for immediate sequestration for a variety of reasons, says John.
A key reason is that many oil reservoirs have potential for incremental oil recovery with CO2 injection that can improve the overall economics for sequestration projects. Geophysical and geochemical modeling after injection will demonstrate methods to monitor injections to make sure CO2 stays in the ground. John, following up on the recently retired Norm Warpinski's work in this area, will be joined by Dave Aldridge, Bruce Engler (both 6116), and Jim Krumhansl (6118).
Peter Kobos (6010), has been at work developing a high-level computer model to analyze physical, economic, and policy requirements needed to understand carbon sequestration in the region. “We've got a prototype right now,” says Peter. “We're trying to integrate all the information so that both experts and interested parties can understand how we assess a project. We've got geologists, regulators, academics, and people from industry all involved. The model is a way we can see all of the issues quickly and address them in an integrated way.”Model components
Len Malczynski (6115) is also a key player on the modeling project, developing much of the software that will tie large amounts of data from the Partnership team to the model itself. Several workshops developed the model’s key parts. An example is screening criteria for underground storage of CO2. A team of geologists helped identify what would be needed to create a successful storage reservoir, Peter said.
Identifying sources of CO2 and how they would flow to these reservoirs, or “sinks,” and some of the economics associated with the project were other model components. Currently, the team is working with a New Mexico test case, but Peter plans to expand it to a regional scope for the future.
“Our goal is an integrated assessment of the costs of the components parts,” says David. “If someone proposes to site a power plant, they will know the costs of carbon sequestration going in. They can look at the infrastructure availability to connect the plant to the sequestration options to help determine the best place to put the plant.”
The concept of carbon avoidance by industry is already being practiced in Europe and is catching on in the US, says David. Colorado has a $9/ton tax credit for CO2 avoided. “This is something that can tip the scale on the type of power plant you might build,” he says. More than 1,000 kilometers of pipeline cross northwestern New Mexico to eastern New Mexico and Texas, moving 25 tons of CO2 a year across the state. This is equivalent to the carbon emissions of about five million people. “We inject and we move it now. The question is how much of a solution is it to the overall problem?”
An estimated 30 percent of US carbon emissions come from power plants and other large “point sources,” like industrial furnaces and refineries. Given the fact that fossil fuels are likely to remain the mainstay for energy production well into this century, most scientists believe that the exploration of “carbon sequestration,” or terrestrial storage of carbon, merits attention.
This is what the President’s Committee of Advisors on Science and Technology reported in a study of 21st-century challenges facing the nation. “A much larger science-based CO2 sequestration program should be developed,” the committee reported. “This is very high-risk, long-term R&D that will not be undertaken by industry alone without strong incentives or regulation, although industry experience and capabilities will be very useful.”
Another approach to the problem is to reduce the carbon emissions from those working power plants through more efficient combustion processes. Sandia, through its Combustion Research Facility at Livermore, Calif., has a role in this approach as well. Sandia researcher Chris Shaddix (8367) and his colleagues are at work on concepts to allow coal-fired plants to burn cleaner, reducing CO2 at the point of emissions instead of storing it. (See stories on clean-burning coal in the next issue of the Lab News.) -- Will Keener
Sandia researchers Ken S. Chen (1514) and Mike Hickner (6245) are working hand in hand to understand key phenomena that control hydrogen-fueled PEM (proton exchange membrane or polymer electrolyte membrane) fuel cells. Ken is developing computational models to describe the phenomena while Mike is performing physical experimentation.
Their work is internally funded through a three-year Laboratory Directed Research and Development (LDRD) grant to tackle several key technical challenges. Proper water management and performance degradation or durability must be addressed before PEM fuel cells can be used to routinely power automobiles and homes.
“A natural byproduct of using hydrogen and oxygen to produce electricity in a PEM fuel cell is water [with waste heat being the other],” Ken, project principal investigator, says. “One challenge is maintaining the proper amount of water in a PEM fuel cell. Sufficient water in the membrane is needed to maintain its conductivity, whereas too much liquid water can result in flooding the cathode gas diffusion layer, which prevents reactant oxygen from reaching catalytic sites and causes performance deterioration.”
The work being done by Ken and Mike is leading to a better understanding of a couple of important areas, including how liquid water is produced, transported, and removed efficiently in PEM fuel cells and how PEM fuel cell performance degrades. A better understanding is key in finding ways to maintain the cells’ long-term performance during normal and harsh (e.g., freezing) conditions and improve their durability.
The close teaming between Ken’s modeling and Mike’s experimental efforts has been quite helpful in meeting project objectives.
“Our approach in combining computational modeling with experiments is unique,” Ken says. “Typically, Mike would perform discovery experiments to gain physical insights. I would then develop a model to describe the observation or data that Mike has obtained. Mike would perform further experiments so I can validate the model I have developed.”
Mike says they’ve obtained some “nice feedback” between the experiments and analyses. The intent is to build a computational tool that can be used in designing fuel cells, eliminating the need to do experiments on every single part of them.
“We want to have all the small pieces worked out in the modeling process so we can concentrate on the larger issues with experiments,” he says.
Ken has been using GOMA, a Sandia-developed multidimensional and multi-physics finite-element computer code, as the basic platform to develop 2-D performance models for PEM fuel cells. With the assistance of Nathan Siegel (6218), he is also exploring the development of quasi-3D PEM fuel cell models using FLUENT, a commercial computational fluid dynamic computer code. Ken emphasizes that the focus of this LDRD project is on understanding the key phenomena using experimental means and computational models, both simplified and multidimensional.
Joel Lash, manager of Multiphase Transport Processes Dept. 1514, concurs. “Sandia’s state-of-the-art multi-physics codes, like GOMA, form the backbone from which simplified phenomena- centric models can be developed to explore complex behavior, such as occurs in operating PEM fuel cells,” he says.
For the past couple of years Ken and Mike have focused mainly on liquid water transport, developing a PEM fuel cell model that can be employed to simulate a fuel cell’s performance, and performing diagnostic tests on fuel cells for phenomena discovery and model validation. Next, Ken says, they will tackle the key technical issues of performance degradation or durability, including performance degradation under normal operating conditions and under freezing operating conditions.
To date, the team — with contributions from Chris Cornelius (6245), David Ingersoll (2521), David Noble (1512), and Nathan Siegel (6218), as well as collaborations with Professor Chao-Yang Wang of Penn State University and researchers at the National Institute of Standards and Technology — has reported portions of its work in three refereed publications, four proceedings papers, and half a dozen technical presentations.
“People are taking notice of our work, and we are at the leading edge of understanding liquid water transport and removal in PEM fuel cells and becoming an important player in the PEM fuel cell research community,” Mike says. “Our validation method is new and exciting and leading us to learn some things not well known previously.”
Bruce Kelley, project manager for the PEM Fuel Cell LDRD and manager of Chemical Biological Systems Dept. 6245, says the project was developed specifically to leverage Sandia’s capabilities in multiphysics modeling and membrane materials to develop broader capabilities with applicability to fuel cells and other related technology areas. In doing so, Bruce says, “We have attracted significant industrial interest in the work, which is key to attracting DOE and other programmatic funding.”
A hydrogen-fueled polymer electrolyte membrane (PEM) fuel cell uses hydrogen and oxygen to generate electricity by an electrochemical process in which electrons are produced in the anodic hydrogen-oxidation reaction and consumed in the cathodic oxygen-reduction reaction. A single fuel cell consists of the MEA (membrane electrode assembly), the anode and cathode GDLs (gas diffusion layers), and GFCs (gas flow channels).
The MEA is the heart of the fuel cell, which is fabricated by sandwiching the polymer electrolyte membrane (e.g., Nafion) between two electrodes. They are composed of conductive carbon support, catalytic platinum particles, and polymer electrolyte binder. Carbon papers or woven carbon cloths are typically used as GDLs. The GFCs are usually etched out of graphite or metal materials. To achieve the desired voltages, single cells are connected in series to produce a fuel cell stack.
In an operating PEM fuel cell, humidified hydrogen is fed to the anode GFCs whereas humidified air is forced through the cathode GFCs. Hydrogen and oxygen are then transported through the respective GDLs. Electrons produced in the anode are conducted through the electrical load to the cathode where they are consumed; protons from the hydrogen oxidation reaction are transported through the membrane. This movement of electrons is an electrical current that can be used to power an automobile or a home. Water and heat are generated in the cathodic oxygen-reduction reaction. The waste heat generated is mostly attributed to the efficiency loss (more specifically, loss due to various over-potentials) in converting chemical energy to electricity.
By Nancy Garcia
A role-playing simulation that enables emergency response officials to see how their decisions might play out in an event may be more important for bioterrorism than for other terrorist
scenarios because it involves release of a biological agent whose effects may take days or weeks to appear, researcher Lynn Yang (8114) said during a briefing of California site managers earlier this month.
The BioDAC (Biological Decision Analysis Center), simulates a release of anthrax or smallpox in an urban area (San Diego County). It was developed by Sandia researchers through the BioNet Program. Funded by the Department of Homeland Security and executed by the Department of Defense, BioNet was a year-long $23 million program to integrate civilian and military biodefense capabilities to facilitate the generation of a unified consequence management plan for a bioterrorism event. This includes jointly detecting and characterizing an event, leading to early phases of the response.
Sandia provided systems modeling and analysis of the population, medical response infrastructure, detection system, and key assets. Role-playing exercises help participants develop countermeasures and responses to an incident.
Ben Wu (8124) presented along with Lynn. He said there are approximately 100,000 military personnel at major Navy facilities in San Diego County, and timing of decisions reflects military mission priorities.
On the other hand, Lynn indicated that the civilian public health officers tend to be fairly conservative in part because of potential risks associated with responses like prophylaxis.
The first indication of an attack may be picked up as an anthrax reading on an air sampling detector in a civilian environmental monitoring system called BioWatch. The positive reading must be confirmed as a true positive. If confirmed, various decision alternatives need to be considered, including when (or if) the public should be notified and when (or if) antibiotic prophylaxis should be distributed.
Although the Navy would make decisions independently, both sides now appreciate the inherent interdependencies of their responses.
Using BioDAC, Navy and public health role players each had distinct views, including maps and other data showing resources and conditions impacting their role. The underlying scenario was visible to the analyst, which showed what Lynn called a “huge attack.” Antibiotic treatment can reduce the impact of the attack if administered in time.
Since the area has about 80,000 tourists a day, she said a scenario involving exposure to smallpox, which is contagious, would be more complicated. The disease would potentially spread beyond the county because large numbers of exposed travelers would leave the area before being aware of being infected.The simulation allows role players to identify and fill gaps in their concepts of operations. It is being evaluated as a possible tool for operations support and training. Insights gleaned from BioDAC contributed to ongoing discussions of the National Bio-Monitoring Architecture. -- Nancy Garcia