Researchers discover astonishing behavior of water confined in carbon nanotubes

November 28, 2016 by David L. Chandler, Massachusetts Institute of Technology
A team at MIT has found an unexpected discovery about water: Inside the tiniest of spaces — in carbon nanotubes whose inner dimensions are not much bigger than a few water molecules — water can freeze solid even at high temperatures that would normally set it boiling. The finding might lead to new applications such as ice-filled wires. Credit: Courtesy of the researchers

It's a well-known fact that water, at sea level, starts to boil at a temperature of 212 degrees Fahrenheit, or 100 degrees Celsius. And scientists have long observed that when water is confined in very small spaces, its boiling and freezing points can change a bit, usually dropping by around 10 C or so.

But now, a team at MIT has found a completely unexpected set of changes: Inside the tiniest of spaces—in carbon nanotubes whose inner dimensions are not much bigger than a few molecules—water can freeze solid even at high temperatures that would normally set it boiling.

The discovery illustrates how even very familiar materials can drastically change their behavior when trapped inside structures measured in nanometers, or billionths of a meter. And the finding might lead to new applications—such as, essentially, ice-filled wires—that take advantage of the unique electrical and thermal properties of ice while remaining stable at .

The results are being reported today in the journal Nature Nanotechnology, in a paper by Michael Strano, the Carbon P. Dubbs Professor in Chemical Engineering at MIT; postdoc Kumar Agrawal; and three others.

"If you confine a fluid to a nanocavity, you can actually distort its phase behavior," Strano says, referring to how and when the substance changes between solid, liquid, and gas phases. Such effects were expected, but the enormous magnitude of the change, and its direction (raising rather than lowering the freezing point), were a complete surprise: In one of the team's tests, the water solidified at a temperature of 105 C or more. (The exact temperature is hard to determine, but 105 C was considered the minimum value in this test; the actual temperature could have been as high as 151 C.)

"The effect is much greater than anyone had anticipated," Strano says.

It turns out that the way water's behavior changes inside the tiny carbon nanotubes—structures the shape of a soda straw, made entirely of carbon atoms but only a few nanometers in diameter—depends crucially on the exact diameter of the tubes. "These are really the smallest pipes you could think of," Strano says. In the experiments, the nanotubes were left open at both ends, with reservoirs of water at each opening.

Even the difference between nanotubes 1.05 nanometers and 1.06 nanometers across made a difference of tens of degrees in the apparent , the researchers found. Such extreme differences were completely unexpected. "All bets are off when you get really small," Strano says. "It's really an unexplored space."

In earlier efforts to understand how water and other fluids would behave when confined to such small spaces, "there were some simulations that showed really contradictory results," he says. Part of the reason for that is many teams weren't able to measure the exact sizes of their carbon nanotubes so precisely, not realizing that such small differences could produce such different outcomes.

In fact, it's surprising that water even enters into these tiny tubes in the first place, Strano says: Carbon nanotubes are thought to be hydrophobic, or water-repelling, so water molecules should have a hard time getting inside. The fact that they do gain entry remains a bit of a mystery, he says.

Strano and his team used highly sensitive imaging systems, using a technique called vibrational spectroscopy, that could track the movement of water inside the nanotubes, thus making its behavior subject to detailed measurement for the first time.

The team can detect not only the presence of water in the tube, but also its phase, he says: "We can tell if it's vapor or liquid, and we can tell if it's in a stiff phase." While the water definitely goes into a solid phase, the team avoids calling it "ice" because that term implies a certain kind of crystalline structure, which they haven't yet been able to show conclusively exists in these confined spaces. "It's not necessarily ice, but it's an ice-like phase," Strano says.

Because this solid water doesn't melt until well above the normal boiling point of water, it should remain perfectly stable indefinitely under room-temperature conditions. That makes it potentially a useful material for a variety of possible applications, he says. For example, it should be possible to make "ice wires" that would be among the best carriers known for protons, because water conducts protons at least 10 times more readily than typical conductive materials. "This gives us very stable water wires, at room temperature," he says.

Explore further: Evidence for solid-liquid critical points of water in carbon nanotubes

More information: Kumar Varoon Agrawal et al, Observation of extreme phase transition temperatures of water confined inside isolated carbon nanotubes, Nature Nanotechnology (2016). DOI: 10.1038/nnano.2016.254

Related Stories

Nanobionic spinach plants can detect explosives

October 31, 2016

Spinach is no longer just a superfood: By embedding leaves with carbon nanotubes, MIT engineers have transformed spinach plants into sensors that can detect explosives and wirelessly relay that information to a handheld device ...

Water, water—the two types of liquid water

November 10, 2016

There are two types of liquid water, according to research carried out by an international scientific collaboration. This new peculiarity adds to the growing list of strange phenomena in what we imagine is a simple substance. ...

Recommended for you

Cellular microRNA detection with miRacles

March 26, 2019

MicroRNAs (miRNAs) are short noncoding regulatory RNAs that can repress gene expression post-transcriptionally and are therefore increasingly used as biomarkers of disease. Detecting miRNAs can be arduous and expensive as ...

What happened before the Big Bang?

March 26, 2019

A team of scientists has proposed a powerful new test for inflation, the theory that the universe dramatically expanded in size in a fleeting fraction of a second right after the Big Bang. Their goal is to give insight into ...

Probiotic bacteria evolve inside mice's GI tracts

March 26, 2019

Probiotics—which are living bacteria taken to promote digestive health—can evolve once inside the body and have the potential to become less effective and sometimes even harmful, according to a new study from Washington ...


Adjust slider to filter visible comments by rank

Display comments: newest first

1 / 5 (2) Nov 28, 2016
This suggests that the nanotubes apply structure to the water molecules, which isn't too strange, I suppose. It suggests that the nanotubes cause entropy to decrease, which means entropy must increase somewhere else. I wonder where the entropy increases?
3.7 / 5 (6) Nov 28, 2016
It resembles polywater....
3.7 / 5 (6) Nov 28, 2016
Or, even better, ice-nine.
Whydening Gyre
3 / 5 (2) Nov 28, 2016
It is curious. I think some sort of electron re-alignment is being performed...
Whydening Gyre
3.7 / 5 (3) Nov 28, 2016
It is curious. I think some sort of electron re-alignment is being performed...

That would more likely mean entropic characteristics are possibly being "focused".
Oops. meant to just edit that last comment, not quote...
3 / 5 (2) Nov 28, 2016
Perhaps larger diameter nanotubes could preserve biological samples indefinitely at room temperature in a solid state.
1 / 5 (1) Nov 28, 2016
The breakdown of the assumed laws of physics in extreme regimes was anticipated by the philosophical outlook of Bernhard Riemann. Such anomalous behavior at these sorts of boundaries should be a priority for research.
4.5 / 5 (4) Nov 29, 2016
Perhaps larger diameter nanotubes could preserve biological samples indefinitely at room temperature in a solid state.

Biological samples are way too big (a single cell is about 4-5 orders of magnitude bigger than the diameter of these nanotubes)

It is curious. I think some sort of electron re-alignment is being performed...

A water molecule is on the order of 2-3 Angstroem (i.e. in a nanotube of 1.5 nanometer diameter you could fit 4-5 of these end to end accross the diameter). At that scale there should be certain, favored, geometric configurations in which the molecules get stuck - which is what I think they refer to when they call it an 'ice-like phase'. With very little variation of the diameter it makes sense to me that the optimal packing geometry can change quite significantly.
Whydening Gyre
5 / 5 (3) Nov 29, 2016
Perhaps larger diameter nanotubes could preserve biological samples indefinitely at room temperature in a solid state.

The freeze phase temp drops the larger the nanotube, so unlikely...
4.7 / 5 (6) Nov 29, 2016
Perhaps larger diameter nanotubes could preserve biological samples indefinitely at room temperature in a solid state.

The freeze phase temp drops the larger the nanotube, so unlikely...

In any case it wouldn't freeze (preserve) the sample. The temperature would still be high enough for stuff to move around in the cell (i.e. the water in the cell itself would not be affected. Only on the outside it would be embedded in an ice-like matrix.

This is not 'freezing' by making stuff cold. This is ordering stuff in a fixed geometry (that's why they call it ice-like and not ice)
1 / 5 (2) Nov 29, 2016
"is curious. I think some sort of electron re-alignment is being performed..."

Theory, The Double electron bond of each carbon atom in a nanotube configuration would align electrons in such a way as to create a "Möbius strip" effect circulating electromagnetic impulses in a spiral, alternating from outer to inner circumference. This would increase the entropy on the inner circumference but would decrease it on the outer producing a two-dimensional phase loop where all the electrons would be influenced to spin in the same direction allowing a bipolar characteristic to occur. Electromagnetic repulsion of the carbon nanotubes exterior would have produced a hydrophobic effect while the interior would have the opposite effect and attracted the water molecules. The configuration would also set an equilibrium within the phase loop that would define and stabilize the temperature and the entropic behavior defined by the diameter of the tube.
5 / 5 (1) Nov 30, 2016
It resembles polywater....

Resembles? It sounds *exactly* like polywater.
1 / 5 (1) Nov 30, 2016
These results confirm a theory that I have proposed in comments to many Physorg articles over more than 5 years. My theory is based on Art Winfree's theory of coupled periodic oscillators. To learn more about his theory, read Steve Strogatz and Ian Stewart, Coupled Oscillators and Biological Synchronization, Scientific American, Dec. 1994 (copy available free online at an site).

Art Winfree was a self-described bio-mathematician. Circa 1966, he developed the math and applied it to biology (synchronization of heart cells, for example). He did not apply his theory to physics. I do. About 300 years earlier, Christian Huygens anticipated Winfree somewhat when he noticed that two pendulum clocks gradually "coupled" the periodic oscillations of their pendula in exactly anti-synchronous fashion.

Winfree's theory is simple. In a coherent system of periodic oscillators, the phases of their oscillations are coordinated in simple, precise ways,, cycle after cycle.
1 / 5 (1) Nov 30, 2016
My Macksb Winfree theory applies equally to normal phases of water (solid, liquid, gas) and to the highly unusual case described in the above article. Both involve coherent self-organization of periodic oscillators with non-linear transitions between their phases.

What do I mean by periodic oscillators? Take electrons. They have two primary periodic oscillations: orbit and spin. Electromagnetic theories, both classical (Maxwell) and quantum (QED), illustrate this point. Electromagnetic radiation, for example, has an electric field oscillating periodically in a (let's say vertical) plane and a magnetic field oscillating exactly in phase (synchronous) but in the orthogonal plane (horizontal). The two cycles are exactly coordinated.

In his two comments above, Antialias is correct in highlighting "optimal packing geometry" and "fixed geometry" in relation to 4 or 5 water molecules across the diameter of the tube. My Macksb Winfree theory explains why.

1 / 5 (4) Nov 30, 2016
Proton wires?

What about all my old electron-based equipment?
1 / 5 (1) Nov 30, 2016
Now for the details, continuing on my two comments above.

Let's compare a 4 water molecule system with a 5 water molecule system. How would the "optimal packing geometry" for a 4 molecule wide system differ from such packing geometry for a 5 molecule wide system? (I use the terminology of Antialias.)

Winfree's theory says that a coherent two oscillator system will have only 2 possible patterns: synchronous or exactly antisynchronous (one-half out of phase in its cycle compared to the other--180 degrees conceptually in a 360 degree cycle). A three oscillator system can be synchronous or 1/3 out of phase (120 degrees for A, 240 degrees for B, 360 degrees for C) and one or two other configurations. A four oscillator system can be synchronous or 1/4 out of phase or it may act like two pairs of 2 oscillator systems. See illustrations in the Scientific American article to which I referred above.

There are additional coherent Winfree patterns for a 5 oscillator system.
1 / 5 (1) Nov 30, 2016
So from Winfree's examples of specific coherent oscillator patterns, we learn that the number of oscillators in a coherent system of oscillators has a big impact on the stable (coherent) patterns or geometry Into which the system can pack itself. (I prefer "self-organize" to "pack," but both terms work.)

Side note: Mother Nature keeps things simple. Winfree's simple patterns, for 2 to about 7 oscillators, will explain the patterns underlying coherent systems of billions of identical oscillators. Insects are a 6 oscillator system--their legs move in Winfree patterns. Millipede legs move like a wave: think 360 degrees divided by X, where X is the number of millipede legs. Same rule as 1/3, 1/4 and so on.

In this particular nanotube example, the Winfree math is particularly easy. Just small numbers.

It is likely, of course, that some of these nanotube cases will have only one possible Winfree pattern. That phase will seem "stiff" or "solid."

1 / 5 (1) Dec 04, 2016
In my second comment above (of four), I mentioned electromagnetic radiation as a good example of two coupled periodic oscillators.

Another example is neutrino oscillation. See the Wikipedia article by that name--especially the color illustrations. There are three types of neutrinos. The periods of their oscillations are coupled in complex ways.

Jan 01, 2017
This comment has been removed by a moderator.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.