A friendly gang of chemists with a penchant for puns, working in a rabbits’ warren of a building at Los Alamos, is in pursuit of a better way to make high-performance lightweight materials. They are developing two new kinds of foam: one is firm like a surfboard, the other is floppy like a yoga mat, and both are light, strong, versatile, reliable, and affordable. They could replace certain conventional materials in the near future.
A foam, in materials-science speak, is a porous polymeric material that has a tailored pore structure. The pores are the key feature—they minimize mass while performing other mechanical functions like insulating, floating, cushioning, and vibration dampening. Foam materials are used ubiquitously in cars, boats, furniture, packaging, shoes, and more. Some are squishy because they must compress, others are rigid because they must not compress, but they all have to be lightweight, long-lasting, chemically inert, and thermally stable.
The production of modern foam materials is plagued with problems. Some foams require complex and specialized manufacturing equipment that takes up a lot of space and can be expensive to source and repair. Some foams require complicated and lengthy processing or suffer from poor process robustness. And many foams include ingredients that are hard to work with, being either toxic, flammable, or both. Polyurethane foam, for example, used in everything from car seats to carpet cushion, starts with isocyanates—a group of organic compounds that are not only toxic and flammable, but are sometimes carcinogenic as well. Even after manufacturing, things like furniture that are made with these chemicals will often off-gas for some time—that’s part of what “new furniture smell” is—and those fumes can cause health problems for consumers.
The Los Alamos scientists have all these pitfalls in their crosshairs. They want to develop high-performance, low-density porous polymeric materials that can be made reliably, quickly, and affordably for national security applications, using standard, easy-to-source equipment and ingredients that are minimally hazardous to the health of the people who make and use them.
“We’re using garden-variety, off-the-shelf supplies to make high-performance materials that are equal to or better than what’s out there already,” says materials scientist Jason Benkoski, who leads the effort.
“It’s a highly interdisciplinary, collaborative project,” adds technologist Eric Martin. “We’ve got chemists, physicists, engineers, materials scientists, machinists—a lot of different people all over the Lab are working with us on this.”
"We're using garden-variety supplies to make high-performance materials that are equal to or better than what's out there."
There are different ways to make a foam. The most common way is to bubble a gas through a liquid that will solidify and trap the gas in bubbles. This creates spherical voids dispersed through a solid matrix, like Swiss cheese. In this type of foam, the surface is the inside walls of the voids, and is mostly concave. The edges between voids can be weak points where the material may tear or crack. Small tears and cracks can lead to larger ones, compromising the integrity of the foam and shortening its life.
Another common method of making a foam is to join spheres of solid material together. This creates a matrix of joined solid spheres, surrounded by void space, like a bag of marbles that have stuck together. In this type of foam, the surface is the outside of the spheres, and is mostly convex. The potential points of failure occur at the necks that connect neighboring spheres.
To circumvent the weaknesses of both concave and convex morphologies, the Los Alamos team is making its materials with neither. Both of the new materials being developed have, instead, what’s called bicontinuous spinodoid morphology. Bicontinuous means that there are two domains—void and matrix—distributed throughout the material. Spinodoid refers to the stable relationship between the two domains: they randomly intertwine such that they are essentially interchangeable mirror images. In other words, if the void space were to become matrix, and the matrix space were to become void, the overall structure of the material will be unchanged.
Spinodoid morphology is the consequence of spinodal decomposition, a phenomenon whereby a single homogeneous phase spontaneously separates into two distinct phases. This kind of morphology offers several benefits. Chief among them is that spinodoids offer excellent correspondence between modeling and experimentation. This allows the scientists to know with high confidence how the materials will behave under conditions that can’t be tested in the lab, based on extrapolation from testing conditions that can be. Also, this morphology is self-assembling—it is the natural consequence of mixing the ingredients with no further manipulation required. Finally, the mechanical properties of bicontinous spinodoid materials are well conserved throughout the material—there is minimal variation from point to point, which means it will perform as expected with high fidelity.
The two domains randomly intertwine and are essentially interchangeable mirror images.
SWIFT is squishy
The first of the new materials is a silicone foam that goes by the name SWIFT (Silicone and Water In Familiar Templates). It is white, pliable, and rubbery. Unlike memory foam, which springs back slowly after being squished, SWIFT springs back quickly.
SWIFT is made from mixing silicone, water, and a chemical catalyst. The process is a lot like making Hollandaise sauce: a water phase and an oil phase are slowly combined with the help of an emulsifier. The ingredients are mixed in a turquoise stand mixer, affectionately dubbed “Taylor SWIFT,” after which the mixture vaguely resembles facial moisturizer. After a brief degassing to remove any air bubbles—whipped-in air bubbles are not how this kind of foam forms—the mixture gets scooped, poured, or injected into a mold. A few hours later, the SWIFT is removed from its mold. It feels a bit like a gel-sole shoe insert and a bit like a kitchen sponge, at once both squishy and solid.
Samples of SWIFT ride on a wheely cart called “The SWIFTmobile 2000,” down the hall and around the corner to be subjected to a battery of tests. Imaging equipment like electron microscopes and x-ray computerized tomography scanners reveal the samples’ inner structures. The main feature under scrutiny is the material’s porosity—or percent overall void space—which largely determines its function. Both the size and distribution of pores contribute to the material’s porosity, with higher porosity making for more squishability.
“The higher the porosity, the more compliant the SWIFT will be,” explains chemist Martin Oltmanns. “Which just means it’s more compressible. We can tailor SWIFT’s physical properties, like compliance and weight, by changing its porosity.”
The way the scientists control SWIFT’s porosity is by changing the concentrations of starting materials and varying the mixing time. The porosity and the stiffness of the silicone matrix can be varied independently; the stiffness can vary by about a factor of four, and the porosity can vary by more than a factor of two. The team is essentially writing a recipe book for SWIFT which will include precise protocols for achieving just about any combination.
The emulsification of silicone and water is what makes the spinodoid morphology spontaneously form. As the silicone polymerizes in the mold, it goes from face-cream consistency to something more like a soaked sponge. Once the SWIFT is removed from the mold, the water evaporates, leaving behind a solid silicone matrix with void space where the water was. The more water added to the mixture initially, the more void space left behind, so the higher the porosity in the final product. Similarly, the longer the mixture is mixed, the finer-grained the bicontinuous spinodoid morphology becomes. So, if two SWIFT samples with, say, 50 percent porosity, are mixed for different amounts of time, one will have larger matrix and void spaces, the other will have smaller matrix and void spaces, but their overall porosities will be equal.
Then there is polymerization time. Different applications may require slow or fast polymerization, which is controlled by the concentration of the catalyst. A spray-on SWIFT, for example, would need to start in a fluid state, then quickly polymerize, so it would require a comparatively high concentration of catalyst. Spray-on SWIFT is still hypothetical, but if the team determines there’s a need for it, they’ll find a way to make it.
The bubbles provide structure and strength without adding weight.
The team is also studying the effects of temperature variation. They want SWIFT to be able to be made anywhere in the country, not just Los Alamos, so the effect of ambient temperature is important to include in their recipe book. A thorough understanding of all these variables will allow for precise tuning of mechanical properties. “It’s all about knowing which knobs to turn to get the right product,” says Martin.
“We’re also learning what matters the most and what matters the least,” adds chemist Mary Knaak. “We need to know which variables are forgiving and which aren’t.”
Knaak, Oltmanns, and Martin crank out batches of SWIFT, carefully noting every weight, temperature, time, and volume. They usually make SWIFT tortillas—flat, floppy, and disk shaped—by smooshing the SWIFT between two steel plates while it polymerizes, because it’s easiest for subsequent analysis. But for instances when a tortilla won’t do and a specific shape is needed, they have a bank of 3D printers to produce specialty molds.
SWIFT could be used in just about any circumstance where a lightweight squishy material is needed. Like cushioning layers between rigid materials, as in body armor for example, or softish coatings on the outsides of stuff, as in various kinds of sports and fitness equipment.
“It could even be used as a combat dressing,” says team member Rebecca Stevens, who has a military background. “It could be infused with antimicrobial medicine, so you could slap it right onto a wound in the field.”
“SWIFT is so darn easy to make,” Stevens chuckles. And she would know, because she makes ELF, the other of the two new foams, and it’s a different beast entirely.
ELF is firm
ELF stands for Extremely Light Foam. It is white, rigid, and strong. It looks like plaster of Paris but it weighs less. ELF is more finicky than SWIFT: it’s more sensitive to temperature, it takes longer to make, and it still involves hazardous starting materials. The goal is to cut the time and toxicity down.
ELF also has bicontinuous spinodoid morphology, and its formation follows the same broad strokes as SWIFT. A two-part epoxy is mixed with a porogen—a chemical that does the same job that water does for SWIFT—then the epoxy hardens and the porogen evaporates, leaving behind a solid epoxy matrix with void space where the porogen had been.
But ELF has something that SWIFT doesn’t have: tiny hollow glass spheres called “bubbles” dispersed throughout the material. The bubbles have an average diameter of 65 micrometers, so that 400 bubbles lined up would span one inch. These don’t create void space; the void space is what’s vacated when the porogen evaporates. The bubbles’ job is lightweighting—they take up a lot of space, providing structure and strength without adding much weight. They also make it easier to tailor the material: how many bubbles are added, how big they are, and what they are made of, are all akin to knobs that can be turned to fine-tune the mechanical properties of ELF. The inclusion of bubbles makes ELF a syntactic foam, a class of materials with an ordered structure that is provided by the inclusion of hollow particles. Syntactic foams feature high strength and low density. They were initially developed for marine applications and are now used in aerospace and ground transportation as well.
Two of the current goals for ELF are to reduce the time it takes to make and to use nontoxic starting materials. But these are tough to reconcile with each other because ingredients that dry quickly, like volatile solvents, also tend to be more toxic than less volatile ones.
Once the epoxy, porogen, and bubbles are mixed and molded, it can take a long time for the porogen to evaporate completely. Small pieces take days, but larger pieces can take over a month. The team wants to switch from an organic solvent to water to have a more environmentally friendly drying process like SWIFT, but drying time is still a problem.
"You solve one problem and a different one turns up. But, hey, that's science!"
Mechanical drying methods may be the trick for making water work. Stevens already uses ovens, but even in the oven it can take weeks, so she is looking at other approaches. The team is presently waiting on a basket centrifuge (and running an in-house poll on what to name it: SPINston Churchill? Carpal TURNer?). It will work much like the spin cycle of a washing machine, by using centrifugal force to pull the water out. Also like the spin cycle, it will greatly reduce the subsequent work that has to be performed by the dryer.
Another challenge that the ELF team is facing is scaling up. “It’s a stepwise process to get ELF to large sizes,” explains Martin. “It’s actually a great example of science in action: you make something, it looks great, then you make it in bulk and bigger, and a new issue appears. You think you may have solved one problem but a different one turns up. Then you discuss the new challenges with your team and develop a path forward. It’s very iterative. But, hey, that’s science!”