Supercomputer simulations reveal entropy roots of water penetration of CNTs

A recent issue of PNAS had an intriguing article that discusses the interplay among carbon nanotubes, water molecules’ hydrogen bonding and entropy, and its worth noting that the Caltech researchers involved in this work say the result might be improved water filtration.

In brief, various groups of investigators have noticed that water tends to flow into CNTs, but according to a Caltech news release, “no one has managed to explain why, at the molecular level, a stable liquid would want to confine itself to such a small area.”

It turns out the behavior (actually various behaviors, depending on the diameters of the CNTs) is largely driven by entropy considerations, i.e., the increase in entropy when H₂O enters the tubes trumps the energy losses incurred by breaking some of waters hydrogen bonds. When these conditions are met, previously stable water can flow spontaneously into the tubes.

The investigators at Caltech’s Materials and Process Simulation Center considered 10 different diameters of CNTs (0.8-2.7 nm), and discovered that three CNT size-dependent flow patterns emerged —

• CNT diameters 0.8–1.0 nm: To fit, water molecules line up one at a time, essentially creating vapor-like conditions for the water, and thus more freedom of motion.
• CNT diameters of 1.1–1.2 nm: There is just enough room for small stacks of water molecules, like crystallized water, and in this size range, bonding—not entropy—is the driving force for the penetration of the water.
• CNT diameters of 1.4–2.7 nm: There is too much room to retain ice-like alignment, and although still confined, the waters “sample a larger configurational space” and exists in a state where some of the hydrogen bonds are broken but it also begins to behave as liquid water.

So, how did lead investigators Tod Pascal and William Goddard figure this out? By developing a suite of simulation algorithms and lots of computing. Actually, that last part about lots of computing isn’t true: Pascal and Goddard’s group replaced molecule-by-molecule supercomputer calculations with what they call an efficient “two-phase thermodynamic model.” In the release, Goddard notes, “The old methods took eight years of computer processing time to arrive at the same entropies that we’re now getting in 36 hours.”

Goddard says their work establishes a valuable theoretical basis for understanding water transport through CNTs, and for nanostructures and nanofluidics in general. He suggests this could lead to some practical applications using nanofiltration and that a possible route to desalination could be to make supermolecules with pores the same diameters as the CNTs. Goddard says a polymer, for example, could made that would “suck” water out of solution.