A CLASSIC experiment beloved of scientifically inclined children is to cover a magnet with a piece of paper and sprinkle iron filings onto the paper. This reveals the field lines that connect the magnet’s north and south poles. Try something similar with some of the new types of magnets now being made using additive manufacturing (3D printing), and a rather different image might appear. Unlike the simple bars and horseshoes of children’s magnets, the 3D-printed variety can be made in all manner of shapes. Their fields can thus be tailored into patterns far more complex than a simple north-south alignment.
These unconventional magnets have huge value in the design and performance of many products that rely on magnetic components: from hospital body-scanners to audio speakers, and from hard disks to wind turbines. In particular, anything that involves an electric motor or a generator also uses magnets. A modern car, for instance, contains a hundred or more electric motors of various sorts, to open and close the windows, adjust the seats, run the heating and, increasingly, to turn the wheels. All require magnets to make them work. The unconventional shapes needed to generate the complex magnetic fields they need to do their jobs properly can, though, be difficult to make.
The other difference between a modern magnet and a childhood one is its composition. Chances are, the magnet under the paper in the school-lab experiment is, like the filings on top, made of iron. The most powerful commercial magnets, by contrast, contain elements known as rare earths. These metals, particularly neodymium, samarium and dysprosium, are not actually all that rare. But they are rarely found in deposits rich enough to be worth mining, so their availability is limited and their prices can be high. Any process that is parsimonious in their use would thus be a boon to industry.
Little by little
At the moment, rare-earth magnets are made in one of two ways. The first is by sintering the required materials together using heat, pressure or both, to create a solid from a mass of powder. The resulting block is then cut and sliced into pieces of the required shape. The second method is to mix the magnetic materials with a polymer, and then shape the mixture by injection moulding to make what is known as a bonded magnet.
In principle, either of these processes might be adapted to the methods of 3D printing. In practice, most such experiments at the moment make bonded magnets. Sintered 3D printing, an established technique, uses a laser or electron beam to heat and melt the powder to be sintered, but the different components of rare-earth-based magnetic materials (the most common is neodymium-iron-boron, or NdFeB) often have wildly different melting points, making sintered printing hard to pull off.
Ways of printing bonded magnets are, however, evolving rapidly, as two recent papers show. Dieter Süss and his colleagues at the Vienna University of Technology, in Austria, have demonstrated a way of printing bonded magnets that resembles the plastic-filament printers many hobbyists use. In this case the filament contains 45-65% by volume of magnetic granules. As the filament is melted, it is extruded by the printer to build a shape up layer by layer. This permits the production of magnets far more complex than injection moulding can turn out, as the team report in Applied Physics Letters. In this case the granules start out in an unmagnetised state, but placing the printed object into a strong magnetic field of the required geometry converts it into a permanent magnet.
Dr Süss’s process, the paper claims, allows new magnet designs to be created on a computer and produced rapidly, with a precision of well under one millimetre. That opens up new possibilities, such as using different materials within a single magnet to create areas of strong and weak magnetism. This could be useful in certain types of sensors.
Parans Paranthaman and his colleagues at Oak Ridge National Laboratory, in Tennessee, meanwhile, have adopted a different technique. They start with pellets containing a blend of 65% NdFeB and 35% nylon. These are then melted and extruded by the laboratory’s Big Area Additive Manufacturing (BAAM) machine. Among other things, BAAM has been used in the past to print car bodies from a mixture of carbon fibre and plastic. In their analysis in Scientific Reports, Dr Paranthaman’s team report that 3D-printed magnets not only retained the magnetism of the materials they were made from, but performed better, in many ways, than those made by injection moulding from similar materials.
Dr Paranthaman says that, with further work, the process should truly outperform injection moulding, especially for making prototypes and short-run customised designs. To change the configuration of a product being made by injection moulding requires expensive retooling. With a 3D printer a software tweak will suffice. 3D printing can be slow, it is true. Dr Paranthaman’s first set of magnets were made with the BAAM nozzle depositing material at a speed of 2½ cm (one inch) a second. But he expects that this could eventually be increased to a metre a second.
Dr Paranthaman and his colleagues are also investigating how to print sintered magnets. In some cases these are more desirable than bonded magnets because they are more powerful (though they are also more brittle and prone to corrosion, and the process of slicing and dicing a sintered block into useful products can waste as much as half of the material in it). Though they are cagey about the details, the team aspire to get around the melting-points problem by spraying a jet of materials onto the surface being built up rather than melting layers of powder. And Dr Paranthaman certainly does not lack ambition. He hopes that, one day, his team will be able to print a steel stator (the stationary part of an electric motor) complete with its rare-earth permanent magnets all in one go.