Chipping Away at Clathrates
New model sorts through thousands of possible conformations of molecular scaffolding
Results: An ideal alternative to imported oil would be a fuel source that is domestically available and environmentally friendly. Large amounts of hydrogen and natural gas (see sidebar) are stored as guests inside clathrate hydrates, a special form of crystalline water found on the ocean's floor and in the earth's permafrost. However, before the guest gases can be used as fuel, significant technical challenges must be solved, among them understanding how clathrates work as scaffolds on the molecular level. Scientists are a step closer, thanks to a computer model built by scientists at the Russian Academy of Sciences and Pacific Northwest National Laboratory.
To make things complicated when modeling those scaffolds, clathrate hydrates can theoretically form many thousands of conformations depending on the arrangement of the hydrogen atoms while obeying the Bernal-Fowler ice rules. Different conformations can behave differently. For example, one conformation could react well to heat, gently releasing its trapped hydrogen. Another might not. The new computer model can quickly sort through the 30,026 possible arrangements of the smallest building block of clathrate hydrates, the pentagonal dodecahedron (aka "soccer ball") cluster of 20 water molecules. In just seconds, the model returned the few most likely energetically stable configurations, the structures of which were further refined on a supercomputer.
Why does it matter? Clathrates pose two possible solutions to the nation's energy crunch. First, clathrates provide a vast resource of natural gas, if we circumvent the technical difficulties of how to extract it. Second, these same scaffolds could hold the key to storing and using hydrogen to power cars and trucks. Several classes of clathrate hydrates meet the Department of Energy's current requirements for hydrogen storage compounds.
Methods: The new model screens quickly through the possible configurations of the water cluster by calculating the energy necessary to hold each collection of water molecules together. Some shapes require very large amounts of energy to make. These are discarded. Other shapes need little energy to build. These are most likely the shapes that would actually exist. These are sorted to the top and provided to the researchers.
The model refined the initial list of 30,026 possible arrangements to just 64, which are most likely to be the energetically stable ones. The team then analyzed these configurations and performed detailed electronic structure calculations using the supercomputer at the Department of Energy's EMSL, a national scientific user facility at PNNL. Each of those calculations took approximately 40 hours to run on 256 processors.
The new model allowed scientists to quickly sort through tens of thousands of conformations and devote their supercomputer run time to the most promising options. In addition, the model identified a new configuration, which happens to be the global minimum of the cluster, that was missed in earlier studies.
What's next: The researchers are looking to better understand how guest molecules are held inside the host clathrate network. They are also planning to study larger clathrates, formed by putting together the smaller building blocks to understand how guest molecules such as hydrogen can be stored inside the scaffolding.
Acknowledgments: DOE's Office of Basic Energy Sciences, Chemical Sciences, Biosciences and Geosciences Division funded the research of Dr. Fanourgakis and Dr. Xantheas. The Russian Academy of Sciences funded the research of Dr. Kirov.
The research was done using computational resources at DOE's EMSL, a national scientific user facility at PNNL.
This work supports PNNL's mission to strengthen U.S. scientific foundations for innovation by developing tools and understanding required to control chemical and physical processes in complex multiphase environments.
Reference: Kirov MV, GS Fanourgakis, and SS Xantheas. 2008. "Identifying the Most Stable Networks in Polyhedral Water Clusters." Chemical Physics Letters 461(2008):180-188. Artwork from this research was featured on the cover of the journal and selected as a Frontiers Article.