GM will work with HRL Laboratories (co-owned by GM and Boeing); Dynalloy, Inc., a Tustin, CA manufacturer of shape memory alloys (Flexinol) specially made to be used as actuators; and the GM/University of Michigan Smart Materials Collaborative Research Lab to develop new low-hysteresis SMA materials and the prototype system.

When a stretched SMA wire is heated, it shrinks back to its pre-stretched length, and when it cools back down, it becomes less stiff and can revert to the original shape. This behavior can theoretically be used to enable a heat engine: a mechanism that is capable of converting heat energy to mechanical or electrical energy.

There was a surge of interest in the potential for SMA heat engines beginning in the 1970s, but the few devices that have been built were too large and too inefficient to make them worthwhile, Aase said.

SMA materials instead have been used as actuators (Dynalloy’s specialty). GM researchers have published a series of papers on using SMA in applications such as latches. The SMA heat engine for waste heat recovery is something the researchers have been talking about for less than a year, Aase said.

We have not written a paper on this particular idea. It had been sort of talked about, but when the ARPA-E solicitation came out [in February], it crystallized our thinking. We should go after this. It’s the kind of thing that is a little bit far out, and since we have so many short term actuator-type projects available to us where we would have a quick return on investment for the company, it’s harder to divert resources to a project that is much higher risk and longer term.

We did some calculations to estimate what we could get out of it. Whether it was theoretically viable, whether we can heat and cool it in an effective manner to cause this two-way action is yet to be proven. This is the kind of high-risk, high-reward activity we thought fit very well with the ARPA-E kind of missions.

—Jan Aase

Key to the success of the project will be developing a material for the particular temperature range of the application, and also to develop a low hysteresis material. SMAs tend to exhibit hysteresis, which would not be good in this application, Aase noted. “It is tolerable as an actuator, but not as a heat engine application.”

From a strictly thermodynamic point of view the SMA thermomechanical waste heat recovery system would not be as efficient as a thermoelectric generator such as the one GM is also developing, Aase said. However, he noted:

Per unit mass, we believe it will be very effective. And in a transportation application, the mass efficiency is very important.

The project at this point is fairly wide open. The team has not yet determined the cooling mechanism. While the simplest thing to do would be to tunnel air past the device, the researchers may also consider taking part of the engine coolant, because liquid has better heat transfer characteristics than air does. “The cooling is the tricky part of the heat cycle,” Aase says.

The material trick is to get the low hysteresis material, and the packaging and design is what we [the collaborative team] will work on together. We assume there will be durability and other issues that come up. We have to go over rollers [in our concept]..bending stresses tend to be rather high. Up to now, all the applications I know of have been with straight pieces of wire. And of course there is also the designing of the heat exchanger portion of it and the mechanical packaging.

—Jan Aase

The pay-off for a successful development could be significant. In a hybrid system, the electrical energy could be used to charge the battery. In a conventional engine, this could perhaps even replace the alternator without any load on the engine, Aase said.

This award is significant for the gains in energy efficiency it could bring, and because it signifies how GM is doing business though collaboration and partnership.

The days are gone when we would do this kind of ground breaking work on our own. We need to continue to find ways to combine our deep technical knowledge with others who can help take our ideas from concept to commercialization.

—Alan Taub, GM vice president of global R&D