Machine — Heal Thyself

Stephen DeAngelis

May 27, 2008

One desired feature of a resilient system is the ability to heal itself. Some animals escape from danger by shedding their tails once in a predator’s grasp. This strategy would only work once if nature hadn’t equipped them with the ability to regrow the lost appendage. Sharks wouldn’t be able to use their slash and tear dining technique if new teeth weren’t lined up to replace those lost during feeding. Although humans can’t spontaneously regenerate lost limbs, they are capable of healing small wounds and recovering from most diseases. The search for self-healing inanimate systems is ongoing but not new. Designers for tire and fuel bladder manufacturers, for example, have managed to create systems that seal themselves when punctured. New materials are being developed that are capable of repairing themselves when damaged [“A healing balm,” The Economist, 8 March 2008 print edition].

“One of the differences between animals and machines is that animal bodies can repair a lot of the damage that a cruel and hostile world inflicts on them. A machine, by contrast, has to wait for someone to come and fix it. But that may change if researchers in the field of self-repairing materials have their way. Two groups in particular—one in America and one in Britain—are trying to create composite materials that mend themselves if they get cracked, in much the same way that an animal’s broken bone heals itself. The difference is that these materials will heal in minutes rather than months.”

It doesn’t take much imagination to understand how such materials could be used in all sorts of settings from aviation to automobiles. The cost of such material will like be cost prohibitive for most day-to-day uses, but they could prove cost effective when used in high risk, high cost system components.

“Such self-healing composites may take a while to enter everyday use. But if they can be made reliably they will be welcome in high-stress applications that are difficult to inspect regularly (the blades of wind turbines, for example) or are critical to safety (such as the doors and window-frames of aircraft).”

The approach being researched by the American team centers around embedding tiny capsules of “medicine” that are released when damage occurs.

“Jeffrey Moore and his colleagues at the University of Illinois are working on the problem by adding extra components to their composites. Like most such materials, these composites consist of fibres (in this instance, carbon fibres) embedded in a plastic matrix (an epoxy resin). The main extra component added by Dr Moore is a sprinkling of tiny capsules containing a chemical called dicyclopentadiene. If the composite cracks, the capsules near the crack break open and release the dicyclopentadiene molecules, which link together to form another type of plastic that binds the crack together and thus heals the material. To start with, Dr Moore had to nurse this process along by adding a second extra component—a catalyst based on ruthenium. This worked well in the laboratory, but ruthenium is too expensive for mass deployment. However, when he was playing with solvents that might be added to the system to speed the transfer of the dicyclopentadiene to the cracks it is intended to heal, he found a solvent that encouraged the process to work without the ruthenium catalyst. Alas, the solvent Dr Moore hit on, chlorobenzene, is pretty nasty stuff (it is used, for example, in the manufacture of DDT). But he has since found a suitable alternative that turns out to be even better. The chlorobenzene process restored only 80% of a material’s original toughness. The new solvents restore it completely.”

The British approach also encapsulates healing substances within their composite material, but in a very different manner than the U.S. team.

“Ian Bond and his colleagues at the University of Bristol’s department of aerospace engineering are taking a slightly different approach. They use glass fibres rather than carbon fibres in their composite and, instead of adding capsules, they have put the healing molecules into the fibres themselves. The molecules in question are the two ingredients of epoxy resin. Half the fibres contain one ingredient and half contain the other. A crack in the material breaks the fibres, releasing the ingredients which react, form more epoxy, and thus mend the crack. The advantage of this approach is that it retains the basic fibre-plus-matrix structure of the material. Adding capsules changes that and risks weakening it. The disadvantage is that capsules are easier (and therefore cheaper) to make than hollow, fluid-filled fibres.”

Like analysts who raise questions about an injured athlete returning to play in a big game, those who must rely on the effectiveness of critical self-healing composites are likely to raise questions about whether the system is as sound (i.e., healthy) as it was before the injury occurred. Unlike an athlete, who can tell those around him how he feels, composite materials don’t currently have a way of notifying operators they have recovered from an injury. Researchers are working on that problem as well.

“Whichever system is adopted (and both might be, for different applications), two further things are needed. One is a way of checking that a component really has healed. The other is a way to top up the healing molecules once some of them have been used. Dr Bond thinks that one way to make healed ‘wounds’ obvious would be to add a bit of colour. A repaired area would, in effect, develop a bruise. Topping up the supply of healing fluid might be done by mimicking another biological system—the network of blood capillaries that supplies living tissues with the stuff they need to thrive. Both Dr Moore and Dr Bond are attempting to borrow from nature this way. If they succeed, the machines of the future will have longer and healthier lives.”

The problem of making healed wounds visible seems easier to solve than the challenge of regenerating the healing mechanism itself. The “blood capillaries” approach, it seems to me, suffers from the fact that the healed wound basically forms a clot in the system preventing the flow of new chemicals. Materials scientists aren’t the only researchers looking to create self-healing systems. IT system designers are also interested in creating systems that can detect and correct defects. These would be software solutions to system disruptions. I don’t know how far in the future such systems will be found, but I’m sure they are coming. When they arrive, the world will be a more resilient place.