Develop Spin New Materials for Low-power, High-speed MRAMs (December 6)
- Develop core material technologies of next-generation MRAM
□ The Ministry of Science and ICT (MSIT) announced that the research teams of Professor Hyunsoo Yang (National University of Singapore, NUS) and Professor Kyung-Jin Lee (Korea University) developed the core material structure of next-generation Magnetic Random Access Memory (MRAM), which can work on low power and high-speed switching.
ㅇ The research was published on online Nature Materials on December 3, 2018 and jointly carried out together with research teams of Professor Awano of Toyota Technological Institute (TTI) of Japan and Professor Qiu of Tongji University of China.
※ title of article: Long spin coherence length and bulk-like spin-orbit torque in ferrimagnetic multilayers.
※ author information: total 17 including J. Yu(NUS), D. Bang(TTI), R. Mishra, R. Ramaswamy (NUS), Jung Hyun Oh, Hyeon-Jong Park, Yunboo Jeong(Korea University), P. V. Thach(TTI), Dong-Kyu Lee, Gyungchoon Go, Seo-Won Lee(Korea University), Y. Wang, S. Shi(NUS), X. Qiu(Tongji University), H. Awano(TTI), Professor Kyung-Jin Lee (corresponding author, Korea University), Professor Hyunsoo Yang (corresponding author, NUS)
□ MRAM has merit in that data can be maintained without external power supply and high-speed motion can be possible, so global semiconductor companies are developing MRAM as next-generation memory in a competitive manner.
ㅇ However, as a way to increase market dominance of MRAM, emerging technologies are required to switch magnetization direction of thick magnetic layer into low current.
ㅇ MRAM motion consists of spin torque incurred by injecting transverse spin current* to magnetic materials. Existing magnetic materials had limit in that thick magnetic layer could not be switched as all of transverse spin current disappears on the surface of materials.
* spin current: In general, electric current refers to a flow of electric charge. Spin current is a phenomenon that involves the movement of ‘spin’, another unique characteristic of electron. Spin current can be free from power loss due to Joule heating, as it can occur without actual movement of electrons.
□ In this research, new material structure, in ferrimagnetic multilayers** which has antiferromagnetic* spin arrangement in atom units and it has been proved in theory and through experiments that transverse spin current is not depleted on the surface of materials and is sustained covering the whole thick layer. This leads to the achievement of spin-charge conversion efficiency that is twenty times higher than existing materials.
* antiferromagnetism: antiferromagnetism refers to spontaneous antiparallel coupling of atomic magnets, whereas ferromagnetism is a phenomenon in which spins of atomic magnets have parallel coupling. Normally, magnetic materials have ferromagnetic properties.
** ferromagnetic multilayers: ferromagnetic multilayers form atomic layers consisting of different elements to cross each other repetitively. In this research, one layer of Cobalt(Co) and another layer of Terbium(Tb) alternate to build its structure and the spins of Cobalt and Terbium have antiparallel coupling at that time.
ㅇ When this new material is adopted to spin torque-based MRAM, which is gaining traction as next-generation memory, this can enhance efficiency of spin torques and have ultrahigh density, as a way to contribute to expanding spin torque MRAM markets.
□ In addition, this material can be adopted to spin-orbit torque MRAM, being developed as future technologies of MRAM. It has high-speed motion and non-volatile traits, which enables it to drastically decrease standby power compared to static random access memory (SRAM) and can be utilized as mobile devices, wearables or IoT memories which require low power.
ㅇ Professor Hyunsoo Yang and Professor Kyung-Jin Lee said, “this research experimentally realizes quantum physics principles which ensures transverse spin current is maintained in magnetic materials to resolve challenges in achieving ultrahigh density of MRAM. This is a good example which illustrates how understanding of basic academic skills can be used in solving critical challenges of applied devices.”
□ The research outcomes were made possible with research support programs of MSIT future material discovery project, senior researcher project, KIST joint research lab project, Samsung Electronics future technology promotion project and research program of the Singapore government.