Self-assembling molecules lend a helping hand
"You don't have to engineer these molecules, and you don't have to manipulate them; they just assemble themselves," says Bobby Sumpter, a chemical physicist who works at ORNL's Center for Nanophase Materials Sciences.
Sumpter and his CNMS colleagues have spent a lot of time developing an array of techniques calculated to harness this natural behavior to enhance energy technologies.
"We can influence how polymers and other materials self-assemble by changing the conditions around them—changing the temperature, adding chemicals or chemical groups to encourage formation of targeted structures or reactions. That's good news for the prospect of discovering both new materials and new applications for existing materials."
The transformational potential of this technology is amplified by the wide-ranging capabilities of research centers such as CNMS. The center's scientists not only can analyze new materials with computer models before they are even created but also can make the materials in the laboratory, then double-check the results of the original computer model using state-of-the-art neutron analysis and electron microscopy facilities. "We're fortunate to have some of the country's best materials scientists here at ORNL," Sumpter says. "Beyond that, we have the computational and characterization infrastructure needed to turn good ideas into reality."
The concept of manipulating nanoscale materials has been around since at least the late 1990s; however researchers didn't have the technical ability to apply those ideas until relatively recently. In the '90s, nanotech was mostly limited to linking polymer molecules together in a line. Today, scientists have uncovered ways of persuading polymers to self-assemble into all kinds of exotic arrangements, creating a range of shapes and mechanical properties in the process. The number of possible combinations generated by these new capabilities is so large that making and testing each one is out of the question.
That's where the laboratory's advanced computing capabilities come into play. Even low-resolution models of these complex materials have to be run on ORNL's Titan supercomputer, the most powerful computer in the world.
"We can model 25 million to 100 million atoms, which is big enough to tell us what we need to know," Sumpter says. "Modeling just nanoseconds' worth of self-assembly is computationally intensive. The process is the same whether we're working with materials for batteries, biomaterials or solar cells. The ability to simulate materials in a digital environment enables researchers to evaluate huge numbers of configurations without having to create them—which would take years. It really accelerates the R&D process."
Sumpter's research group is pursuing a number of nanotech projects, including those described below, that involve various self-assembling materials—all with potential applications in the energy field:
Smaller, lighter batteries—Improving the capacity and durability of batteries is particularly important for electric vehicles and grid storage applications. Nanotech researchers at the laboratory are experimenting with batteries that use solid polymer electrolytes rather than typical liquid electrolytes. This reduces the size of batteries by eliminating the need for separation of liquid electrolytes on the positive and negative sides of the battery. The use of solid electrolytes also addresses a number of problems related to deterioration of the battery's electrodes over time.
"Solid electrolytes reduce both the size and weight of batteries," Sumpter says, "and we have been able to design polymers that conduct ions as well as traditional electrolytes."
When researchers use computer simulation to develop materials such as solid electrolytes, it's an iterative process. If a simulation shows promising properties, researchers make the material and study its characteristics using neutron analysis and electron microscopy. If the results of these studies match the simulation, the material is likely to be explored further, and the accuracy of the computer model will have been validated, increasing its usefulness for modeling similar materials.
Cleaner water—ORNL scientists have looked into the possibility of using durable, flexible polymer membranes for separating liquids, including water filtration applications that remove brine and other impurities.
"The current technology for water filtration is very energy-intensive," Sumpter says. "Water is possibly the most important commodity in the world, so there is a huge demand for energy-efficient ways to purify it. A breakthrough in this area could be transformational. We're investigating physical and chemical filtration processes. Both techniques involve membranes that are activated by the presence of impurities. So far materials like this haven't made the jump to industrial applications, but they offer a lot of promising possibilities."
Smart materials—An oft-cited goal of nanotech research is the development of materials that are "stimuli-responsive," meaning they respond to environmental conditions such as temperature or mechanical strain by displaying new structures or properties.
Smart materials usually have either the ability to sense "problems" within themselves, such as strains or defects that could eventually cause the material to fail, or self-healing capabilities—such as the capacity to self-repair chemical bonds that have been broken by mechanical stress.
"Our group is working on materials that can sense problems," Sumpter says. "Energy science-wise, it would be an advantage if one could prevent or predict catastrophic materials failures, which are rather costly both in terms of energy and, more importantly, life."
His group and their collaborators have been investigating the possibility of creating such materials using carbon nanostructures that look like stacks of ice cream cones. When the stacks are straight, their electrical resistance is high. When they bend in response to an outside force, their resistance drops to near zero.
The interesting thing about these stacks is that the cones are not chemically bonded; they're just slipped inside one another. The ability of this arrangement to vary conductivity in response to strain lends itself to building sensors into the fabric of, perhaps, critical structural components of transportation vehicles (such as airplane wings) and similar structures where measuring stress is important.
Light-related properties—A number of energy technologies would benefit from materials with enhanced light-related properties. Toward this end, CNMS scientists have been experimenting with enhancing polymers to improve their potential for use in light emitting diodes or to respond to changes in light for use in solar cells.
Sumpter notes that one of the challenges of working with bulk quantities of self-assembling materials is that sometimes they don't self-assemble into structures of optimal size. Additionally the structures that these self-assembling materials create may contain defects.
"Defects usually have a negative effect on the properties of the material, and it's hard to get rid of them," Sumpter says. "A single defect, for example, can have a profound effect on a material's ability to conduct electricity or light."
Eliminating defects generally involves investigating why they occur and finding ways to avoid them. One of the most successful workarounds has been the practice of using a defect-free crystalline surface such as copper, gold or silver as a pattern or template for self-assembling molecules. As it turns out, structures that self-assemble on top of this type of template tend to be defect-free as well.
"We learn a lot from developing solutions to these kinds of problems," Sumpter says. "We attempt to tease them apart from the bottom up. It's a step toward understanding what's important in nanoscale systems and what's not."
Point, counterpoint, solution
It's fair to say that pairing experimentation with computation in areas such as polymer self-assembly could shave years off the process of developing new materials and is transforming the field of nanotechnology. Sumpter, who has worked in the field since its earliest days, says the biggest difference between now and 15 years ago is that today researchers have a lot more confidence in the results of computer simulations.
"Computation is a good counterpoint to intuition," he says. "We've had a number of situations where intuition was leading us in one direction, and our computational models were leading us in another. Now we trust the models enough to believe them when they say Option A is better than Option B. Having the computing power needed to create highly detailed material models has reduced the time required for problem-solving by a huge amount, so now the idea of using theory to guide experiment is actually practical."
The potential for groundbreaking research to be conducted at CNMS is enhanced by scientists' access to all the resources necessary to exploit the technology.
"We can handle the full spectrum of R&D activities at the laboratory," Sumpter says. "Our ability to model materials, make them and then measure their properties is what makes ORNL and the Center for Nanophase Materials Sciences unique—and kind of neat, I think." óJim Pearce