Nanoscopic crystals about 1,000 times thinner than a human hair wade through a liquid solution, leisurely consolidating into chains of three or four links here, five or six links there.

Over the course of another few minutes, they form longer and longer chains by latching onto others of similar length. Given enough time, this process of self-assembly will produce a material with tailored properties imbued by its crystals.

As reported April 4 in the journal Nature Communications, an international team led by University of Nebraska-Lincoln engineers Eli and Peter Sutter has become the first to directly image this process in the liquid environments that foster it. The study also presents detailed statistical models that account for various forces driving the self-assembly.

Nanocrystals appeal to scientists and engineers for multiple reasons, not least because changing their shape, size and composition can dramatically alter electronic, magnetic and other properties in the materials they form. They also tend to bond with one another more agreeably than do atoms in conventional materials, offering the sort of a la carte versatility that the latter cannot.

“You want to be able to make these nanocrystal materials on demand,” said Peter Sutter, professor of electrical and computer engineering. “This means that you’d like to understand where and how they form, and also how you could reshape them.

“The ideal way of doing this is by using an imaging technique that works in the liquid environment, that has sufficient resolution to follow single nanocrystals, and that also has a time resolution good enough that the assembly process can be tracked in real time.”

The new study demonstrates that liquid-cell electron microscopy, an imaging technique that transmits high-energy electron beams through very thin layers of liquid, checks all three boxes.

Alternative techniques, such as those that scatter X-rays from these structures, already allowed researchers to precisely measure the structure and particle spacing of 3-D nanomaterials that self-assemble in liquids. Despite this, they ultimately failed to yield the insights that the Sutters and their colleagues were seeking.

“What they cannot tell you is how the crystal comes together in the first place,” said Eli Sutter, professor of mechanical and materials engineering. “Diffraction only gives you a signal that can be interpreted once the crystal has a certain size. This means that the initial stages of the growth of these nanoparticle crystals, and of any reconfiguration processes, are completely missed.”

In contrast, Eli said, the new study represents a first step toward observing how changes in a liquid solution – its pH level, its salinity, its polarity – might reconfigure nanocrystals on the fly. This understanding could herald unprecedented control over their assembly, she said, offering the potential to “really make an adaptive ‘smart material’ out of them.”

The team’s observations also revealed surprises about the self-assembly of octapods, which are branched nanocrystals that resemble eight-pronged jacks. The researchers anticipated that individual octapods would initially link up to form two-crystal chains, which would then lengthen by attaching to other nanocrystals one at a time.

Instead, the crystals seemed to resist pairing up, suggesting that a three-crystal chain is the shortest stable length. And rather than snagging individual octapods, chains seemed to favor linking up with other chains floating in the solution.

The Sutters and their colleagues also discovered that the nanocrystal links enjoyed substantial wiggle room, giving the chains flexibility while migrating through the liquid solution. Contrary to expectations, this flexibility – which introduced additional measures of randomness, or entropy – actually served to further stabilize the chain.

“So basically, by having this additional … freedom of motion, the chain actually fixes itself more strongly than if the octapods were pulled closer together,” Peter said. “This entropy actually plays a big role in the nanocrystal self-assembly.

“You normally associate entropy with disorder, but in this case entropic interactions drive the ordered state.”

These observations informed the compilation of a statistical model that successfully predicts the one-dimensional assembly of octapod crystals. The next challenge: imaging three-dimensional nanocrystal growth in liquids, which could involve building composite images from a series of two-dimensional slices.

The Sutters might also use the newly demonstrated imaging capability to explore dynamics of protein folding and other biological processes occurring on the nanoscale, some of which could be applied toward improving the delivery of pharmaceutical drugs.

“There should be a fair amount of curiosity in understanding things like protein folding in the native solution environment,” Eli said. “These are things that should be interesting to study going forward.”

The Sutters co-authored the Nature Communications study with researchers from Brookhaven National Laboratory, the Italian Institute of Technology and the University of Stuttgart (Germany). The group’s research received support from the U.S. Department of Energy and the European Union.

Engineering Nanomaterials