|
New device materials: why?Lighting, public transport, manufacturing and personal computing - these are central to our modern lives. Unfortunately, right now, light bulbs waste 95% of the electricity we put into them, whereas the AC motors and power supplies used in transport and industry can waste up to 45% and the RAM in personal computers can waste tens of watts even when it isn't used. Given our rising demand for energy but limited fossil fuel supplies, this is a major problem! However, major energy savings can be made by improving just two basic types of electrical device; light-emitting diodes (LEDs) and transistors. In particular, we need much more efficient green LEDs (to be combined with existing red and blue LEDs to produce white light) and we need transistors that can run efficiently at very high powers and frequencies without wasting energy on standby. Such devices could also be shrunk and adapted for use in ultra-high-density computer memory. However, current materials cannot reach the performance needed for these devices, so better materials must be found. What do novel nitrides have to offer? Materials in electronic devices usually have just one main function. For example, gallium nitride works as a semiconductor in blue LEDs and high-power transistors. However, this proposal centres on creating multifunctional nitride-based materials for use in new, improved devices. Currently, some exotic materials can simultaneously act as semiconductors, ferroelectrics (i.e. they have a spontaneous, reversible electric polarization) and as magnets, but most of them are unstable, difficult to manufacture or don't work at room temperature. Instead, existing nitride semiconductors could be modified by adding metals like scandium, which generate tiny distortions in the crystal structure. These materials are particularly exciting because the distortions can produce new ferroelectric and magnetic properties which nobody thought could coexist in the nitrides. At low metal concentrations, the new materials are stable and can emit light of the right colour to replace existing, highly defective active regions in green LEDs. At higher metal concentrations, the distortions line up and the entire crystal structure changes. Such materials could then be used in transistors, where they should produce a thin switchable layer of electrons, giving a very low 'on' resistance without drawing power when 'off'. Alternatively, by detecting the presence or absence of this electron layer, we could take away the transistor 'source' and 'drain' and create dense, stackable arrays of nanometer-sized devices which could provide record-breaking data storage densities. Depending on how the materials' magnetic and electrical properties interact, multiple bits of information might even be stored simultaneously. These new materials are expected to be both robust and compatible with existing nitride processing technology, making them of great practical value. Firstly, however, their fundamental properties must be understood more fully, in order to make the most of the fascinating new possibilities they offer for the energy-efficient devices of the future. This can be done by creating and characterising the most promising materials (starting with the (Sc,In,Ga)N materials system), understanding and controlling their fundamental properties and using this knowledge to design new energy-efficient devices that best exploit these properties. Impact: Better green LEDs could help save up to 80% of the energy we use in lighting. Along with more efficient high-power transistors for industry, transport and communications, this would reduce our dependence on fossil fuels significantly. In the long term, such energy-efficient displays and power supplies could also be combined with ultra-high-density memory to give us smaller, faster, lighter computers with enormous data storage capacities and very long battery lives, benefiting almost every part of society.
|