80

nanotimes News in Brief

Chips // Silicon Chip “Replaces” Rare Earths

R

are earths are an expensive and necessary com- ponent of strong permanent magnets. However,

their use for this purpose can be optimised and thereby reduced. This has been demonstrated in computer simulations by a Special Research Program funded by the Austrian Science Fund FWF.

The results show that such magnets may contain local deformations in the crystal lattice of the ma- terial. These deformations are above all located at the boundary of material grains. According to the calculations of the St. Pölten University of Applied Sciences, the magnetic force of the material is wea- kened in these areas. This could be avoided by op- timising the material structure, which would save resources by reducing the amount of rare earths required.

The team at St. Pölten University studied the ex- act structure of neodymium magnets. In addition to the rare earth element neodymium, the magnets consist of iron and boron. The head of the Industri- al Simulations study course, Prof. Thomas Schrefl, commented on the recent findings: “Our simulations show disturbances in the crystalline structure in neodymium magnets. Such disturbances cause the magnetising direction to change in these areas. In a so-called anisotropic magnet, like the neodymi- um magnet, in which all parts must have the same magnetising direction, this phenomenon weakens the magnet.” The team’s simulations show that such

disturbances in the junctions between individual ma- terial grains occur when three different grains meet. In these triple junctions, a non-magnetic enclosure is formed and the crystal lattice near the enclosure is disturbed. In the same region, a high demagnetising field weakens the magnet further.

The influence of disturbances on the magnet’s beha- viour were found in multiscale simulations that take into account several different dimensions: from the atomistic to the visible range. Conventional simulati- ons were unable to cover this range of size until now. It was the combination of individual numerical com- putational methods, such as fast boundary element methods and tensor grid methods for computing the magnetic fields, which finally made it possible.

The spokesperson for the Special Research Program, Prof. Georg Kresse from the research group Compu- tational Materials Physics at the University of Vienna, explained the aims of the Special Research Program: “We want to describe the correlated movement of electrons more accurately. This electron correlation is mainly responsible for the cohesion of solid-state bodies and molecules. An accurate description is the- refore crucial for precisely predicting the mechanical, electronic and optical properties of materials.”

11-02/03 :: February / March 2011