Modelling negative thermal expansion

Materials World magazine
1 May 2007

The adverse effect of thermal expansion on a range of applications, including bridges and fuel cells, can now be minimised, say scientists at the University of Malta. They have devised a model to predict the impact of heat on a structure’s shape.

‘Thermal expansion can create enormous problems,’ explains lead researcher Dr Joseph Grima. ‘Particularly in situations where a system is required to operate at a wide range of temperatures, you have a contact between two or more materials that expand differently from each other – creating unwanted forces – or when thermal expansion results in the closure of gaps that must be present between electrodes in fuel cells [and] sensors.

‘A significant amount of work has therefore been performed on materials which exhibit anomalous thermal expansion properties, including negative thermal expansion – materials which shrink on heating. We wanted to study systems which can produce similar effects but can also be produced cheaply and on a large scale.’

This research follows work carried out by scientists at the Universities of Cambridge and Exeter, both in the UK. It involves creating a structure from triangles constructed using three rods that are connected to each other by rotating joints. If one side of the triangles is made from a material that responds to rising temperatures more than the other two sides, the structure has been shown to shrink in one direction when heated, becoming shorter in height.

‘Any rigid material – metals, alloys, plastics – can be used in the construction of the systems,’ says Grima. ‘The important thing is to have at least one component that has a different thermal expansion coefficient from the rest.’

To enable better control of thermal properties, the triangles can also be created using three different materials at varying lengths. Grima adds, ‘Through careful control of the materials and dimensions used, the resulting system can be made to exhibit a wide range of thermal expansion coefficients, and thus we can tailor-make structures for use in practical applications.

‘The equations we derived can be used to precisely compute and predict the percentage changes in dimensions of the overall structure per degree of temperature change.’

By virtue of its rigid framework and use of conventional materials, the team argues that the model is suitable for load-bearing functions and could eventually be mass produced cheaply to any size specification, be it for use in construction or a microscopic system. Further research is now focusing on testing the design using finite element simulations and real life examples, such as by incorporating the triangular structure into a softer rubber material.

Grima also works on auxetic materials (negative Poisson’s ratio) that expand when stretched instead of getting thinner, and sees potential for combining the two projects. He says, ‘One of our aims is to produce systems that have both properties simultaneously negative. This is interesting as auxetics exhibit enhanced physical characteristics ranging from increased indentation resistance to improved acoustic damping properties.’


Further information:

Joseph Grima, email: