Computational methods have become key to tackling every aspect of the U.S. energy challenges, from stretching the fossil fuels we primarily rely on today, to developing new energy sources and improving the alternatives we already employ, such as solar and wind power.
To bring down the costs of producing biofuels, researchers need to understand how to break down the tough fibers of plant cellulose into sugars. Molecular-level simulations are helping them understand how enzymes – present in fungi and other organisms – already perform this trick.
These kinds of simulations are also used to understand photosynthesis to build better, more efficient solar cells.
Even with progress in cleaner energy sources, fossil fuel-fired power plants will continue to generate 69 percent of American electricity in 2035, according to the U.S. Energy Information Administration.
Computation is key to developing cleaner fossil-fuel combustion, such as coal gasification and high-efficiency, low pollutant industrial burners.
Atom-by-atom simulations, freely available models, and open databases speed the search for the new materials we need for efficient fuel cells, longer-lasting batteries, and high-tech membranes that can separate carbon dioxide and other harmful gases out of fuel plant exhaust.
In 2012, the President emphasized the importance of scientific computing, databases, and new algorithms to the race for these new materials when he announced the $100 million Materials Genome Initiative.
Scientific Computation in Action
One idea for keeping excess carbon dioxide (CO2) out of the air is to capture it from fossil-fuel burning plants and pump it into saltwater reservoirs deep underground.
To understand the long-term fate of CO2 in such “geologic sequestration” sites, researchers are turning to simulations like this one, created at Lawrence Berkeley National Laboratory. This 3D model shows how some CO2 mixes with brine over time creating long fingers that begin sinking. Unlike earlier models, this one uses a modeling technique that focuses computing power on the most active portons of the simulation, yielding more detailed results.
New computer simulations are able to predict the performance of large wind power plants with greater accuracy.
Design tools are generally effective for basic optimization of wind farm layout, but they can’t simulate with consistent accuracy how wakes propagate or how wind turbines interact with one another to impact efficiency. Researchers at the National Renewable Energy Laboratory are creating more sophisticated models that look at a wind farm as a total system rather than just a collection of wind turbines.