When two look-a-like balls are not always exactly the same

Magneter Energiteknologi Elektromagnetisme

Differences of less than one percent and very small variations in the chemical composition of the nominally same material can have a significant impact on the resulting effect in magnetic cooling. A new research project at DTU will make it possible to take the small but influential impurities into account when dealing with magnetocaloric materials, thereby making magnetic cooling more efficient.

If you examine a batch of identical steel balls for use in ball bearings in detail, you’ll despite the apparently identical loks find countless small differences in mass and chemical composition between individual balls.

Mostly those small differences do not matter to the use of the balls, but once in a while even the smallest differences in chemical composition prove out to be of great importance. Magnetocaloric materials are materials that change temperature when magnetized by a magnet, and here even small differences can have a major impact if the components are used as active materials in a magnetic refrigeration machine.

"A magnetic refrigeration device is critically dependent on the tailor-made properties of the magnetocaloric materials, but we often find materials that, in theory, look quite reasonable and useful, but when it comes down to practice, they flunk and can’t fulfill their promises," explains Associate Professor Kaspar Kirstein Nielsen, researcher in Magnetic cooling at DTU Energy.

He has just received a grant of 2,592,000 DKK from the Danish Council for Independent Research for the project "Getting the most out of magnetocaloric materials for high efficiency refrigeration". The goal is to develop a model that takes the small impurities and differences in composition of magnetocaloric materials into consideration, making it possible to utilize the materials far better in refrigerators.

"We often find materials that, in theory, look quite reasonable and useful, but when it comes down to practice, they flunk and can’t fulfill their promises"
Kaspar Kirstein Nielsen, Associate Professor and researcher in Magnetic cooling at DTU Energy

Magnetic cooling is based on the fundamental thermodynamic characteristic of magnetic materials, the so-called magnetocaloric effect. When a material is magnetized, it warms up and when the magnetism is later removed, the material cools down. An effect used to create pollution-free cooling using magnetism. The problem occurs if the material is not uniform, if different areas or sub-elements of the same material react differently to the magnetism, stress and/or heat transfer. Experience shows that there can be significant differences between the theoretical predictions of a sample material and the measured properties.

"Even slight differences and small variations in the chemical composition can have a significant impact on the effect of the entire system. So I can never assume that my material is absolutely perfect, because it's not really. Therefore, we must take into account the impurities, as even small impurities can affect the magnetocaloric properties a lot. "

Kaspar Kirstein Nielsen and DTU have initiated collaboration with research groups under Professor Lesley Cohen of the Imperial College of London (ICL), UK, and Professor Vitalij Pecharsky of Ames Laboratory, Iowa, USA, on a research project in which a series of supposedly uniform samples will be exposed to a wide range of tests from stress tests over measurements of the sample's heat conductivity to x-ray diffraction.

"Most measurements are made here at DTU Energy, but our partners have access to equipment we haven’t so a PhD in the project will travel around with the samples, testing the exact same samples using different machines and techniques here, in England and in the United States. During the process a certain amount of wear will be imposed on the samples, such as a volume change of up to one percent when a material is used for magnetocaloric tests, but that is all part of the test as changes occur at different times in different places in a material if it is a bit unclean. And that is precisely what we want to measure and make a model over, enabling us to take the impurities into account in the experiments and increasing our understanding of the basic properties of the materials in the process," says Kaspar Kirstein Nielsen.