Magnetoresistive devices using highly spin-polarized materials

• Report rights infringement
• published: 31 Jan 2011
• views: 1574

[Keio Spintronics Network - Shirai Laboratory, Tohoku University] The Masafumi Shirai Laboratory at Tohoku University's　Research Institute of Electrical Communication is doing theoretical research on electrical conductivity in magnetoresistive devices. The aim is to achieve very functional spintronics devices using highly spin-polarized materials. Q. "The hard disk drives in today's PCs, and future non-volatile magnetic memory, utilize tunneling magnetoresistive devices, which combine magnetic and insulating materials. In our group, we theoretically calculate what sort of materials could be used to improve performance in the future, and how those materials should be combined to improve device performance. And we use these calculations to provide experimentalists with guidelines for fabricating new devices. " To increase circuit integration in nonvolatile memory using magnetic materials in the future, an important technology will be alignment of the magnetic moments perpendicular to the plane. For example, if a perpendicularly magnetized material called iron-platinum is combined with magnesium oxide, it's clear from theoretical calculations that the memory reading output isn't very high. But the Shirai Lab has also calculated that a high output can be obtained if a very thin layer of iron, just a few atoms thick, is formed between the iron-platinum and the magnesium oxide. Q. "From our calculations, we can obtain guidelines as to whether the iron should be introduced only on one side of the magnesium oxide, or on both sides, and how many atomic layers thick the iron should be to achieve a very high output. And from these calculations, we predict that the output won't be high unless iron is introduced on both sides of the magnesium oxide. But in fact, we've found that just by adding a layer of iron a few atomic layers thick -- in this calculation, it's actually three atomic layers thick -- we can get a large output, over 1,000 percent, with a high tunnel magnetoresistance. So we hope this will be a guideline for ongoing experimental R&D;. " Nonvolatile memory, where data is preserved even when the power is off, is also expected to be the key to making future electronic devices less power-consumption. And this type of memory also utilizes tunneling magnetoresistive devices. The Shirai Lab is investigating the use of materials with high spin polarization. However, to utilize such materials in spintronics devices, they must be in contact with non-magnetic materials. As a result, the performance of the high spin polarization is degraded around interfaces. Q. "In reality, at an interface, down spin states occur as well as up spins, so the situation isn't ideal. And this degrades the properties of spintronics devices and tunneling magnetoresistive devices. We propose that one way to overcome these problems is to carefully adjust the atomic arrangement at an interface. So we've been able to theoretically suggest material systems that could maintain 100% spin polarization even at the interface. For example, one proposed method is to introduce a layer of chromium and aluminum atoms just one atom thick between the magnesium oxide and the highly spin-polarized material, which is composed of cobalt manganese and silicon." In this way, guidelines based on theoretical calculations play a very important role in developing spintronics devices, which will form the basis for future information technology. Q. "At present, we're satisfied if one in ten of our suggestions works. But we want to do something to improve the success rate. To achieve this, we need to know a lot more about how various properties of a material are related to the arrangement of its atoms or their electronic states. We'd especially like lots of young people to take part in this research, helping to build up new theories and predictive methodologies.