Carmel Majidi, Professor of Mechanical Engineering at Carnegie Mellon University, and his team has developed a material with a unique combination of high electrical and thermal conductivity, with actuation and self-repair capabilities that are unlike any other soft composite.
Tech Briefs: What gave you the initial idea to combine liquid metal microdroplets with liquid crystal elastomers? Did you do it with a clear goal in mind?
Professor Carmel Majidi: We did, this was a public project funded by the U.S. Army Research Office. The original idea was to create, basically a kind of robotic material — a material that did more than just provide structural support, but could also be actuated to change its shape, could sense touch, and process information. The idea was that by combining liquid elastomers (LCEs), which are shape programmable, with liquid metal, which is electrically conductive, you might end up with a composite that has a unique combination of material functions that go beyond just electrical conductivity and mechanical integrity.
The idea of using droplets of liquid metal originated from previous work that my lab had been doing incorporating microdroplets in a fairly wide range of different types of polymers — mostly soft rubbers. What we were finding, was that depending on the concentration of the droplets and certain kinds of processing conditions, we could tailor the electrical and thermal properties of the rubber without altering its intrinsic mechanical properties. We tried a silicone rubber, which is very elastic, compliant, and rubbery. But silicones are electrically and thermally insulating. By adding droplets of liquid metal, we can enhance both the thermal and electrical conductivity without altering the rubber’s elasticity. We can make rubber composites that have thermal conductivity close to metals like stainless steel. We can also engineer them so they’re electrically conductive — not quite as conductive as pure metal, but conductive enough for sensing and digital circuitry. And the inclusions of liquid, even at high concentrations, don’t have too much impact on the mechanical properties of the rubber — it remains very elastic, very soft —softer than skin, and stretchable to five or six times its natural length. That inspired me to think about other sorts of material systems where we want to tune the thermal and electrical properties without sacrificing the mechanical features. Liquid crystals came to mind because they have some really interesting programmable mechanical properties — active dynamic shape and thickness steering properties. And then we thought that by adding liquid metal, because it’s liquid, it’s not going to alter or impair its natural shape-memory properties but would enhance the thermal and electrical enhancing properties.
Tech Briefs: Could you explain how the shape-changing works?
Professor Majidi: The way the liquid crystal elastomer works, is it has a shape memory effect. At room temperature, the material can be very malleable, and you can easily deform it. But when you heat it, the gel undergoes a kind of a phase change of the liquid crystals. That causes the material to spontaneously reconfigure and transition between a deformed shape and the original preprogrammed shape. Temperature is used to cause the phase change to go back to its original shape. So, with liquid crystals, up until we started working with them, to get this type of behavior, you had to apply the heat externally. By adding liquid metal to make the composite electrically conductive, we can simply pass electrical current through it and induce internal heating within the material. That current will then induce the phase change and we’ll see the shape change response. It eliminates the need for using a hair dryer or a heat gun — some external kind of heating source — instead, we can just power it electrically and get the same effect.
Tech Briefs: When you heat it and it changes shape, can you predict what the new shape will be?
Professor Majidi: We can; the shapes are dictated by the alignment of the liquid crystal molecules, which are just like the ones used in the liquid crystal display on a television. These liquid crystal molecules — mesogens — are connected to the polymers in the rubber. They can pull on the rubber and influence it into a completely different shape. We induce the liquid crystal molecules to change their configuration with respect to each other. They then pull on the polymer rubber chain and cause the rubber to contract or deform — we can program that. For this project, we used a UV laser to selectively program the tightly linked structures of the liquid crystal mesogens, so that when we induced the phase changes, we could design the shape that the material would spontaneously deform into. In this manner, we can produce a highly complex shape. In our study, we showed a kind of folded origami shape, a wavy type shape, and something that reversibly transitions from being flat to being spherical. The simplest shape change was a strip that goes from being elongated to contracting to a short length. That change can be used to create a soft actuator — kind a soft artificial muscle.
Tech Briefs: So, you program it once when the material is produced?
Professor Majidi: Yes, that’s correct. When we’re processing the material, that’s when we have the UV laser come in. The UV exposure is very localized over a few 10s of microns. We polymerize the materials a certain way so as to program the shape.
Tech Briefs: I saw a demonstration of the electrical properties in a video of you with Hari Srinivasan on PBS SciTech.
Professor Majidi: Yes, the circuit that self-repairs when I cut the material. That was different, it was silicone rubber, but it’s the same principle. We saw that we could achieve the same thing with the liquid crystal elastomers as well. You have these liquid metal droplets inside of the polymer matrix, and if they’re touching each other, you have electrically conductive pathways. Then if you cut or puncture or damage the conductive trace, the liquid metal around that damaged area will spontaneously rupture and form connections with their neighbors to form new conductive traces around the damaged area. We realized that this property would enable us to produce a damage-responsive material that self-repairs. This was kind of a cool effect we had never seen before with other material composites.
Tech Briefs: Is it very deterministic that when the ruptured droplets are cut, they will connect with the adjacent droplets?
Professor Majidi: What we found was it’s somewhat predictable, but it highly depends on the nature of the damage. Light pressure is not be enough to cause new conductive pathways to form, but if there’s really intense pressure or if the material is cut with a scissors or an
X-acto knife, or punctured with a hole puncher, that will induce needed conductive pathways around the damaged area. Whether the overall conductivity of the soft rubbery circuit increases or slightly decreases from what it was before highly depends on the type of damage. With puncturing you see one effect, cutting with scissors or tearing by hand, you see a different effect. It’s not fully deterministic in that sense, but it is repeatable, so for certain types of applied pressure or for certain types of mechanical damage, we do see a somewhat repeatable response.
Tech Briefs: I saw in that video you had what looked like a printed circuit, where conductive paths were embedded in the insulating material. How do you produce that?
Professor Majidi: We use a Cricut desktop die-cutting machine. We mount a pen to the motor — and basically get an x-y plotter — that scribes the surface of the material. It allows us to control the depth of indentation so that everywhere the pen presses into the material, we get electrical conductivity.
Tech Briefs: How do you make connections between your traces and a piece of electronics?
Professor Majidi: That’s something we’re still working on. For the past number of years, we’ve been developing ways to form robust connections between the pins on these different packaged electronics with the liquid metal circuits. One way we found that’s been very effective, is to use either a hydrochloric acid vapor or sodium hydroxide vapor. When we expose the liquid metal to the vapor, it removes the oxide skin that exists on the surface of the liquid metal. Removing the oxide allows it to have very high surface tension and wet to other metal surfaces. When we place a packaged electronic chip onto our circuit, the liquid metals will make nominal contact with the pins of the IC chip. Then we blow some of that vapor over the circuit to remove the oxide skin insulating barrier from the liquid metal. That allows it to grab on to the pins of the chip. It’s so powerful that even if you plop it down with tweezers, or position it manually by eye, so that it’s not perfectly positioned, the liquid metal surface tension will grab onto the pin and force it to reorient. That’s true of a lot of soldering, like with more conventional manufacturing of printed circuit boards. It’s just that with our liquid metal, instead of using a flux, we have to use this vapor to achieve wetting.
Tech Briefs: Do you foresee these primarily being used as conductors?
Professor Majidi: Yes. The idea is using the liquid metal we can wire packaged chips together over a large area in a way that is mechanically robust, stretchable, soft enough to be compatible with a person’s body, and also self-healing — resistant to tear and other mechanical damage.
Another area we’re interested in is thermal management of these sorts of soft electronics. When you start populating a lot of microelectronic chips like CPUs and radio transceivers you generate a lot of heat during operation. With the enhanced thermal conductivity of our rubber we can have a thermal interface material, or a kind of rubber coating, that can quickly dissipate and spread the heat out from the electronics.
Tech Briefs: That sounds like a win-win. You get the flexible conductor plus a heat sink.
Professor Majidi: Yes, in fact, in a lot of these applications, the focus is more on the thermal conductivity as opposed to the electrical conductivity. I’m actually calling you right now from one of the spinoff companies from my lab (Arieca LLC) that’s focused on commercializing the thermal conductive rubber — we call it Thubber™. It’s basically silicone rubber with these liquid metal droplets. We’re looking at applications like thermal interface materials in electronics. It’s a very simple concept — at the end of the day, we’re just adding these micro-droplets of liquid metal to existing rubbers.
Tech Briefs: What kind of ambient temperature range can this work over?
Professor Majidi: We just published a paper where we report the properties of these composites over a pretty wide range of temperatures. What we found was that because the droplets are so small, they exhibit a super-cooling property, such that they’re still liquid even at very low temperatures — much lower than the natural melting point of the liquid metal itself. The liquid metal we’re using — gallium indium alloy — has melting point of about 15°C and as we’re cooling it, it has a crystallization point of approximately -10°C. But, shrinking the liquid metal down to microscopic droplets, suppresses the crystallization temperature down to about -70 to -80°C, which is pretty extraordinary. Since there are rubbers that can stay rubbery down to these very low temperatures, we can make liquid metal-rubber composites that can stay rubbery, and soft, and stretchable even at -70°C. And then on the high end, the boiling point of liquid metal is incredibly high, so the limiting factor there is basically what the polymer can support. Many silicone rubbers start to degrade at above 200°C, so that’s typically the upper limit for the composite.
Tech Briefs: What practical applications do you see coming down the road?
Professor Majidi: The one we are most excited about now is the Thubber, which is being commercialized by Arieca. Their primary emphasis is on electronics and semiconductors, using these materials as thermal interfaces between, say, a CPU and a heat exchanger or a fan. Thubber can also be used within packaged electronics because it is soft and can be made into a very thin film. More generally, it can be used in places where it’s difficult to use thermal grease, which are susceptible to pump-out and leaking, or aluminum plates, which are relatively rigid and bulky.
There are many exciting opportunities: aerospace, automotive, even healthcare — basically anywhere you want a rubber that’s good at conducting heat. For example, cooling and heating garments. It can also be used as a highly conductive rubber lining on grommets, gaskets and seals. It can be used as heating and cooling elements in car seats and interiors. These are all places where there’s interest in having materials with high thermal conductivity but are also soft enough that they can interface with the human body or mate two surfaces together in a seamless fashion.
Those are some of the near-term commercial applications. Longer term, the electrically conductive versions have really interesting properties, in that the electrical conductivity doesn’t change much when you stretch the material, which is very unique. Most existing conductive rubbers, once you stretch them to a few times their natural length, their conductivity plummets, whereas these maintain a steady electrical conductivity. This property has been especially interesting to researchers at the Air Force, where they’re developing stretchable antennas and radio transmission lines. These materials behave very well even at high frequencies, into the gigahertz range, and we’re currently working with our Air Force partners to do this characterization.
Further along, we could see them being used as circuit wiring for highly stretchable garments, wearables, or inflatable structures.
An edited version of this interview appeared in the January Issue of Tech Briefs.