(Left) Infrared image of a fiber-reinforced composite during self-healing of an internal delamination via thermal remending. (Right) Optical images of 3D printed thermoplastic healant (blue overlay) on a carbon-fiber textile. (Image: Jason Patrick and Alexander Snyder)
Tech Briefs: What first led you to the idea for this project?

Professor Jason Patrick:I first started working on self-healing materials, namely polymers and composites, during my PhD at the University of Illinois Urbana Champaign. I was developing microvascular-based self-healing systems — bio-inspired systems where we would create vascular networks, much like blood vessels in living organisms — inside synthetic structural materials. The approach relied on two-part liquid agents, reactive chemistries, contained within the internal vasculature. Fracture damage to the material would release the sequestered healing agents, which would come into contact and polymerize in situ to repair the damage. However, I kept running into hurdles with this two-part vascular approach. Mainly repair inconsistency and limited repeatability — less than 10 healing cycles. We would get buildup of polymer, scabbing/scarring, which would obstruct subsequent flow. Cross contamination also occurred, blocking off vascular delivery conduits. Thus, repeatability and reliability of those materials were a challenge. And healing took about two days to obtain sufficient recovery. This all got me thinking that there's got to be a better way to do this. That was really the catalyst for this new self-healing strategy.

Tech Briefs: In your new composite, is the healing agent separate from the bonding agent?

Patrick: Typically, when we think of a fiber-reinforced composite, it's got two primary phases, one being the reinforcement — the fibrous phase. The other component being an encompassing matrix. For this system, we're reinforcing an epoxy matrix with glass and carbon fiber.

We're working with laminated composites: layers of reinforcement stacked in a particular configuration and infused with an epoxy matrix, a thermoset, which cures and solidifies, resulting in a strong and lightweight structural component. However, such laminated materials are inherently prone to separation of the matrix from a reinforcement layer — interlaminar delamination.

Our self-healing system relies on the addition of two other materials. One being the healing agent, a thermoplastic, which we 3D print directly onto the primary reinforcement, glass or carbon. And we also include resistive heaters, which are actually layered structural reinforcement. The interior layer contains an electrically conductive network of carbon whiskers. Both exterior layers, sandwiched around the network of carbon whiskers, are glass fiber reinforcement. The glass electrically insulates the heater, so you don't experience shorting or anything of that nature, particularly in carbon-fiber reinforced composites, which are conductive. We always incorporate two heaters and place them symmetrically about the composite midplane to avoid warping and bending/twisting coupling effects.

Tech Briefs: At what temperature does the healing agent melt?

Patrick: Let me first describe the healing agent and then I can explain more about temperature. The healing agent is a thermoplastic, so when it's heated, it softens and melts and can be reformed. It's the type of thermoplastic adhesive that's traditionally used to seal food packaging. So, if you've ever struggled with a potato chip bag or a cereal bag and found it very hard to open — resistant to fracture — we're using a similar adhesive to those sealants. Specifically, we’re using poly(ethylene-co-methacrylic acid) or EMAA for short. Some attractive features are, it's a readily available commodity polymer, non-toxic, and as I mentioned, notoriously tough to separate.

The melting point is around 85 °C. Self-healing takes place at a little higher temperature than that — somewhere above melting but below the glass transition temperature of the epoxy matrix. If you're above the glass transition temperature (Tg) of the epoxy, it's going to transition from a glassy to a rubbery state — it will lose stiffness and get softer. So, to ensure we maintain the structural integrity of the fiber composite, healing occurs below the Tg of the epoxy thermoset but above the melting temperature of EMAA. That window is somewhere between 100 °C and 130 °C. It depends on how much recovery you want to achieve. At higher temperatures, the healing agent will have lower viscosity, resulting in better infiltration into small cracks and confined regions. Though the closer you get to the epoxy matrix glass transition, you will sacrifice a bit in terms of structural rigidity, although it's fairly minimal at the temperatures we’ve been using.

Tech Briefs: What about direct sunlight on the structure — could that initiate the melting?

Patrick: For direct sunlight, unless you have a kind of a focusing lens, it's probably not high enough to melt and achieve healing. You would need some additional energy, which might come from electrical power, or maybe other sources to get the healing to occur. So, I don't think direct sunlight is necessarily going to do that, although direct sunlight on Earth is one thing — in space, it’s another. So, it really depends on the application and the environment.

Tech Briefs: Would the amount of time that you apply the heating vary based on the environment?

Patrick: We haven't done a lot of studies looking at different time durations. In our lab at room temperature, we found that 15 minutes is enough for sufficient healing to occur, and the cooldown period to return to ambient conditions was 30 minutes or so.

This is markedly faster than the previously developed vascular technology that took several days, 48+ hours.

Tech Briefs: Could you describe the conductive heater layers in more detail.

Patrick: The electrically conductive portion of the heater is a veil of carbon whiskers. They're short fibers, not continuous, and they're distributed in such a way that they overlap and interconnect — or percolate — and produce an electrically conductive pathway. These conductive layers are tuned for a certain resistance so that we can achieve resistive (Joule) heating at a suitable electrical input power. Tech Briefs: How do you tune it?

Patrick: Tuning depends on the distribution of the carbon whiskers, how dense a network of fibers you have, how random the orientation is and how much interconnection. We're working with a company that has developed these heaters for other applications; we don't actually make them ourselves.

Tech Briefs: So, you order a specific resistance range?

Patrick: Yes, but we are now looking into developing custom heaters, with modified material makeup and geometries.

Tech Briefs: I guess the advantage of the carbon whiskers is that the heating is more uniform.

Patrick: Yes, heating is fairly uniform, and the heaters are embedded. The heaters are thin and structural, so there's good compatibility with the rest of the laminated composite. Resistive heating, especially with metals, has been around for a long time. But generally, you don't have good bonding between metal wires in epoxy. So, these thin layered heaters are a composite-compatible solution for in situ resistive heating.

Tech Briefs: What kind of power are you talking about?

Patrick: That's a good question. For the laboratory samples we tested, the input power was somewhere between 12 and 15 Watts. Obviously, power input is going to scale with size. But what we envision is that you don't necessarily need to put these heater layers everywhere. They can be placed in high-stress or failure-critical locations, plus heat can travel if thermally conductive pathways are established. We’re also looking at ways to reduce power consumption through new heater materials and geometry modifications. But we don't envision, say, an entire airplane wing or wind turbine blade having a heater throughout — just in certain locations.

Tech Briefs: What size samples have you used in your laboratory tests?

Patrick: Our fracture samples were roughly 150 millimeters long by 25 millimeters wide and 5 millimeters thick. So, test coupon geometries, but the heater was spanning the entire region, and we propagated a crack that was about 50 millimeters long by 25 millimeters wide — a significantly sized defect.

Tech Briefs: What are you planning to use as the power source for the heater?

Patrick: We've done a little research on that. We envision utilizing available on-board power for applications like aircraft and automobiles. We're also thinking about emerging structural power sources, which are multifunctional composites that have energy harvesting and power generation capabilities built into the material itself. There are a few groups around the world that are actually creating batteries out of structural composites.

Tech Briefs: So in in the real world, how would one decide when to initiate a repair?

Patrick: That’s a great question. At this stage, with what we've developed, we envision healing as a maintenance cycle that the structural system would go through. For instance, with an aircraft after a specified number of flight hours, you would apply the heating and automatically repair any internal damage. This could be done grounded, or in-flight if below the Tg of the composite matrix. Or you can think about a leveraging a system that has heaters embedded for deicing. Thus, every time you activate the deicing system it also repairs the structure.

But the nice part is that you don't have to locate the damage. Wherever there's a crack, the molten healing agent is going to flow into the crack and repair it. In fact, the healing agent is a bit specialized — there are chemical reactions that occur between the EMAA healing agent and the epoxy matrix, which produce water vapor and pressurize the healing agent thereby forcing it into cracks. So, it's ideal from the standpoint that these internal fissures will get filled and repaired anytime there is sufficient heating and a location for the healing agent to flow into.

Another thing that we're exploring right now is self-sensing — which is very close to being worked out. We’re developing sensing technologies that will not only detect damage in order to trigger the healing to occur, but also monitor the repair process it as it evolves, so that we can quantify the extent of mechanical recovery during service. Right now, we have to destructively test samples to measure healing recovery, which is appreciable, between 80-100 percent, depending on system configuration. But we are looking at ways to do that sensing internally, so one would have real-time feedback and a fully autonomous system.

Tech Briefs: How would you do that sensing?

Patrick: Unfortunately, I can’t tell you all those details at this time, but what I can say is that we are going to be able to sense internal damage and healing and link these two phenomena. We're talking about self-sensing in terms of a material level sensor.

Tech Briefs: What do you mean by material level?

Patrick: We're not going to have an embedded electronic device; it's going to be built into the actual composite material platform.

Tech Briefs: I have a feeling this will have important effects in the world.

Patrick: Yes, I think it will lead to a much different mindset in terms of structural practice. We won’t need to have bulky overdesigns or manual inspection to try and locate subsurface damage. In complex fiber-composite materials, you always have a sub-surface defect somewhere that can’t be detected, whether it's from initial manufacturing, or long-term fatigue, or an impact/overload event. The idea is, let's live with defects/damage as long as they can be self-repaired and not allowed to grow to a size-scale that can cause catastrophic failure.

One promising aspect about the healing agent that we 3D-print: it actually increases the resistance to fracture in the first place — upwards of 500 percent increase in the initial resistance to interlaminar fracture. And on top of that, we have the healing capabilities, which can recover up to 100 percent of that already enhanced fracture resistance, and we can do it repeatedly. I think that's the real differentiator here, that we were able to demonstrate 100 cycles, which represents an order of magnitude greater than existing self-healing composite technology, and without exhibiting signs of waning.

The healing is repeatable and it's very reliable — it has consistent fracture recovery. We’re heavily focused on damage-tolerant structures with upfront greater resistance, where post-damage structural properties can be recovered over and over again. That's what we're ultimately after.

Tech Briefs: Any thoughts about the earliest applications?

Patrick: Yes, we've given that quite a bit of thought. Fiber-reinforced composites are ubiquitous in modern-day applications, everything from civilian and military aircraft to energy infrastructure (wind turbine blades), and even sporting goods. So really, any composite infrastructure is a viable candidate. But there's always an economic tradeoff that we have to consider. And so, I think that the most likely early applications will be in the area of safety-critical or difficult-to-access structures. For example, rotating components such as helicopter blades, or wind turbine blades, both of which have been known to delaminate, and can be life-threatening, or at a minimum, a very difficult and costly situation.

We are looking for industry or funding partners to give us an application and say: “Here are the material requirements that we need to meet. Can you do this?” And the nice thing is, because we use 3D printing, we can tailor the geometry. We can also tailor where the heating occurs. We can consider the structural requirements to determine how to minimize any potential impact from these composite modifications. The idea is, whatever the application, we can customize the design for the end-user’s application.

The overarching goal is to increase the reliability of structural composites and also to enhance their resilience, to mitigate maintenance costs and downtime. This will also eliminate waste from material combinations such as thermoset matrix composites, which are inherently difficult to recycle. I don't know if you've ever seen wind turbine blade landfills — they're huge. There's a lot of research currently looking into new recyclable thermoset composites. We're trying to help established materials last longer and improve their overall safety by preventing failures.

A more sustainable world is what we're after here. We need composite materials: they're lightweight and good for energy efficiency. For aircraft, less fuel is needed if you can reduce the weight of the structure, why not also make them last and continue to fly longer?

Tech Briefs: Do you have a sense of the timeline for when we might see actual real-world applications?

Patrick: We already have a lab-scale prototype that works. Going forward, it really depends on how quickly other entities want to adopt this technology by working with us to help meet their requirements. Whether it’s industry or perhaps DoD entities that are interested in exploring further. We have already been contacted by an aerospace company overseas that is interested in testing some of our material in their vehicles.

So, it's hard for me to answer when exactly, but I think certainly within the next few years, maybe even sooner, once we can get some prototypes on real structures, evaluate, and modify as needed. We’re very much open to potential partners approaching us with their desired applications.

We’d also love to work with NASA and send the materials into space. With vascular systems, space environments are challenging — liquid healing agents would likely freeze or be very difficult to deliver in atypical atmospheres. But with this new thermal remending system, we have the ability to tune temperature and drive the thermodynamic healing process in any environment, even in a vacuum or in freezing cold temperatures. Sky, or should I say space, is the limit.