New Heat Pump Stretches Towards Higher Efficiency

A group of researchers at DTU Energy have demonstrated a novel type of heat pump based on materials which heat up when they are stretched. The potential for high efficiency and low materials cost of the device makes it a promising alternative to conventional refrigeration and heating based on vapor compression.

heat pump

A sizable and growing part of the world’s electricity consumption is spent on refrigerators, air conditioners and heat pumps. As the world moves towards a more sustainable energy system there is an increased focus on energy efficiency. The technology behind heat pumps and refrigerators, vapor compression, is mature and highly optimized. However, it also has inherent limitations of its efficiency, and the most widely used refrigerants are greenhouse gases. For this reason a significant amount of research on alternative refrigeration technologies is being done around the world. DTU Energy has been active in this field for more than 10 years, in particular within magnetic refrigeration.

“We are very happy with the performance of our device, as we have only worked on elastocalorics for a couple of years. It could not have been done without the experience gained through many years of work on the related technology of magnetic refrigeration.”

Professor Nini Pryds


The principle behind the new heat pump is really very simple. Pull a rubber band quickly and you will feel it heat up. Release it again and it cools down. What happens is that the long rubber molecules which are randomly curled up in the unstretched state will stretch out and align in response to the pulling. This means that the molecular disorder has decreased – in physics this disorder is called the entropy. If you pull fast enough such that the rubber band does not exchange energy with the surrounding air, the total entropy of the rubber band does not change. This means that the decrease of orientational disorder will be accompanied by an equal increase in the disordered heat motion of the molecules, i.e. the temperature of the rubber band is increased. This so-called elastocaloric effect is reversible: When the rubber band is released, the molecules curl up again, and the temperature decreases. The fact that the effect is reversible makes it possible to create a high-efficiency cooling or heating cycle based on it.

Instead of rubber the researchers used a special alloy of nickel and titanium. It has the property that it can be made to transform reversibly between two different configurations of its atomic lattice by applying a stress. This transformation is associated with a similar transformational entropy as in the stretching of rubber. Thin plates of the alloy were laser-welded together and clamped to a mechanical actuator capable of periodically stretching and releasing the plates. To transport the heat out of the material, water flowed back and forth inside the stack of plates. By careful arrangement of the piping, it is possible to have a unidirectional flow of water outside the stack, but an alternating flow inside. The alternating flow is timed with respect to the stretching and releasing of the plates in such a way that a temperature difference is built up between the upper and lower inlet. It is the first time that this so-called regenerative principle has been used for elastocalorics. It allows a sizable temperature difference to be produced while achieving a high efficiency of the device.

The researchers are encouraged by the initial results: A maximum temperature span of 15.3 °C with a specific heating power of 800 W per kg of the active material, with a coefficient of performance (COP) of 3.5.  The COP measures the amount of heat delivered per unit of work input to the device. Even higher COPs of up to 7 were realised for different operation conditions. Such performance makes it realistic to see applications of the elastocaloric effect for heat pumps and refrigeration.

The device was constructed as a first demonstration of the potential for regenerative elastocaloric heat pumps. There has as yet been no systematic optimization of the design which means that the researchers still see lots of opportunities for improvements. In particular they point out the need of operating two elastocaloric stacks in tandem to utilize the input work most efficiently. The researchers also plan to explore more advanced stack geometries and materials.