IBM’s Virtual Supercomputer to Improve Water Quality
According to a paper published today in the journal Nature Nanotechnology, scientists discovered a phenomenon in which the use of carbon nanotubes, under specific conditions, could potentially lead to more efficient water filtering with less expense and less impact on the environment.
The researchers’ study of nanotubes — material that shows promise for a variety of technologies — was enabled by a virtual supercomputer created by IBM in which volunteers channel the surplus processing power of their computing devices to scientists for use in conducting simulations.
Carbon nanotubes – tiny, hollow structures made of a material related to graphite in pencils – are so small that they may filter out impurities from water flowing through them.
The scientific community initially expected that their narrow diameters would slow the water’s flow. Surprisingly, early experiments hinted that water is not impeded in the expected way as it passes through nanotubes.
To understand why, a distinguished team of international researchers led by scientists at Tsinghua University undertook an unprecedented, massive computational simulation study powered by IBM’s World Community Grid to find out what was behind this surprise.
Prior simulations performed by the scientific community were unable to study the process at realistic water flow rates because that would have required considerably more costly computing power than typically available.
The new simulations were conducted using the massive computing power of IBM’s crowdsourced World Community Grid, which revealed that under certain conditions, the natural, random thermal vibrations of atoms in nanotubes could have a significant effect on water moving through them.
The researchers discovered that these vibrations, called phonons, can actually enhance the rate of water diffusion — a kind of flow — by more than 300%, as a result of reduced friction.
Researchers led by the Center for Nano and Micro Mechanics at Tsinghua University in Beijing performed vast simulations using the donated, surplus processing power of IBM’s World Community Grid, which harnesses three million linked computers from more than 700,000 “citizen-scientist” volunteers worldwide.
The nearly 100 million calculations performed by IBM’s virtual, crowdsourced supercomputer for the Computing For Clean Water project would have cost USD $15 million had they been performed commercially, and would have taken more than 37,000 years had they been performed on a single-processor PC. Instead, the work was completed at no cost to scientists and in a fraction of the time.
With this newfound understanding of the phenomenon, researchers now hope to optimize the nanotubes and apply them to improve water filtration and seawater desalination.
As freshwater reserves dwindle worldwide, an improved and less costly purification process could help quench thirst and grow crops. Nearly one billion people around the world currently lack access to safe drinking water.
The new understanding of this phenomenon may also lead to a better understanding of how chemicals and drugs pass through tiny channels in human cell walls, potentially leading to improvements in medicines.
With further research, it might also be possible to apply these findings to improve a process that creates energy when freshwater and saltwater are mixed, a process known as osmotic power.
International collaborators who contributed to this discovery include researchers from Tsinghua University, University College London, Tel Aviv University, University of Geneva, University of Sydney, Monash University, and Xi’an Jiaotong University.
In the picture above: Go With the Flow: Scientists at Tsinghua University in China use IBM’s World Community Grid to discover the conditions necessary for moving water through carbon nanotubes 300% faster without requiring additional energy. The discovery has implications for more efficient water filtration. Scientists used massive computing power from volunteers to create simulations of water flow at the molecular level with astonishing detail. (Photo Credit: Tsinghua University)