A team led by National Tsing Hua University (NTHU) professors yesterday announced a breakthrough in magnetic random access memory (MRAM) semiconductor research.
MRAM is considered a better memory device than dynamic random access memory (DRAM) and static random access memory (SRAM), as it is faster and more energy-efficient, and its stored information remains intact even when it is powered down, NTHU Department of Materials Science and Engineering professor Lai Chih-huang (賴志煌) said.
MRAM works by manipulating its ferromagnetic and antiferromagnetic layers (classes of magnetic materials), but researchers have been working to make the manipulation more flexible, without having to change the MRAM’s external temperature, Lai said.
After adding a platinum layer under the ferromagnetic and antiferromagnetic layers, the team succeeded in using a spin current caused by the flow of electrons to manipulate their “exchange bias,” a phenomenon occurring in magnetic multilayers where the antiferromagnetic layer “fixes” the ferromagnetic layer, he said.
The team is the world’s first to use a spin current to manipulate the exchange bias, resolving an obstacle that researchers have battled with for the past 60 years, he said.
Understanding spin current is vital for semiconductor development, as it has been pushed to the extreme, he said.
The team patented the technique in Taiwan, the US and China before publishing their findings in a paper titled “Manipulating exchange bias by spin-orbit torque” in the journal Nature Materials on Feb. 18, Lai said.
To verify their experiments, the team also developed techniques to measure temperature variations in microseconds, NTHU Department of Physics professor Lin Hsiu-hau (林秀豪) said.
While most scientists are still debating whether a spin current can manipulate the switching of micromagnets beyond 1 nanometer, the team has demonstrated that manipulations caused by a spin current can persist for more than several nanometers, Lin said.
After submitting their manuscript to the journal, the team spent more than one year convincing the journal’s reviewers that it was a Taiwanese team who achieved the breakthrough, despite limited funding, he added.
Funded by the Ministry of Science and Technology, the team also works with the ministry’s Taiwan Semiconductor Research Institute and other universities, Lai said.
Their technique might be used in the mass production of new MRAM devices in four or five years, Lai said, adding that they are meeting with domestic semiconductor firms to discuss a technology transfer.
RSS