07-06-2021 | | By Robin Mitchell
Recently, researchers have been able to combine Iron and Gallium to create a new magnetostrictive material ideal for use in Magneto-Electric Spin-Orbit electronics. What are Magneto-Electric Spin-Orbit electronics, how have the researchers created the new material, and why is it important?
While trying to increase the number of transistors on a chip is arguably the main goal of all semiconductor foundries, the secondary goal is to try and minimise the power consumed by each transistor. Transistors that consume less energy result in devices that consume less energy, and therefore such devices are more easily integrated into battery-operated devices such as smartphones and tablets.
One method to reduce the amount of power consumed by transistors is to reduce the voltages at which they operate. However, CMOS technology is reaching its limits, and the continuing reduction of transistor gates is having less of an effect on the gate voltage (scaling on gate voltages stops at 0.5V).
Magneto-Electric Spin-Orbit devices could be the solution to this as they can be operated at far lower voltages, can be constructed using similar techniques to modern semiconductors, and are already being actively developed by major companies including Intel.
Magneto-Electric Spin-Orbit devices, or MESO, are electronic devices that utilise the interaction between magnetic fields and the spin of electrons. A MESO device operates in a very similar way to CMOS devices in that it has an input, an output, and a power supply. The input current generates a magnetic field that interacts with a special spin-orbit layer (i.e. affecting the spin of electrons). The degree to which the spin of the electrons is affected determines the current that can flow from the output (hence, the input current controlling the output current). However, the voltage needed for such devices is less than 100mV which is significantly smaller than that of CMOS devices. Furthermore, MESO also offers a technology that can be potentially reduced below the nm scale, and the reduced energy requirements from 300x10-18J to 10x10-18J represents a significant reduction.
Recently, a team of researchers from the University of Michigan announced the development of a new magnetostrictive material that provides major improvements over currently used materials. One of the biggest challenges faced by MESO is the need for rare-earth elements with ideal magnetostrictive properties. The better the properties, the better the switching action of the device, and the less energy needed to use the material. However, finding such materials can be a challenge, and any MESO device that is to become a reality needs to be mass-producible.
Iron and Gallium combined can create a magnetostrictive material whose properties are subpar to rare-earth materials, but their low price makes them ideal for commercialisation. As a result, the research team explored methods to create a material with improved magnetostrictive properties with the hope that it can be used as an economical alternative.
Before the researchers work, the creation of an iron-gallium magnetostrictive material would fail when trying to use over 19% gallium. Beyond this value, the material undergoes a phase transition and results in capping the magnetostrictive capabilities. The researchers, however, have taken advantage of thin-films and epitaxial growth to maintain the metastable iron-gallium structure at gallium concentrations up to 30%. Furthermore, the research team also took advantage of a second phase-change that occurs in Iron-Gallium at 30% Ga to create a magnetostrictive material whose magnetostrictive properties are 20x that of 19% Iron-Gallium.
The ability to replace CMOS with MESO would lead to lower-powered devices, lower energy consumption, and potential faster devices. Furthermore, the simplistic construction of MESO devices could also allow for sub-nm devices, leading to an explosion of technological development. In order to realise this, materials with good magnetostrictive properties need to be identified and using commonly found elements such as Iron and Gallium greatly helps with this.
The device produced by the researchers is far from ideal, and its large size (measured in microns), means that it cannot be used in a commercial device. However, the team is working with Intel who has taken a special interest in the project. The collaboration is looking to reduce the size of the device so that it may be a candidate for replacing CMOS in the future.
CMOS has been the workhorse of electronics for the past 40 years. Still, now that we are approaching the physical limits of CMOS, it’s time to look towards the next generation of technologies that will allow for smaller devices that consume less power.