Your browser does not support JavaScript!

:::

中文版    
 

 

NTHU Research Team at the Cutting Edge of Memory Research and Development
From mobile phones to computers, wearable devices, the Internet of Things, and smart city operations, data processing and storage are the core of digital technology. At present these functions are mainly carried out using dynamic random access memory (DRAM), but this consumes power and can’t easily be further reduced in size. However, a research team led by Professor Lai Chih-huang of the Department of Materials Science and Engineering and Professor Lin Hsiu-hau of the Department of Physics has made a revolutionary breakthrough in the development of the next generation of digital memory known as magnetoresistive random access memory (MRAM), which allows for faster reading and writing, uses less power, and retains data even when there is a power outage.
 
Their research has been published in a recent issue of Nature Materials, and has had an impact factor of 39.2. With the support of the Semiconductor Moonshot Project of the Ministry of Science and Technology, their research is expected to make a major contribution to the domestic memory industry.
 
Reducing size and energy consumption with MRAM
 
Professor Lai explained that DRAM uses the electrical charge of electrons; the presence of the charge is represented as “1,” its absence as “0,” and maintaining this information requires a continuous power supply. As for size, DRAM originally conformed to Moore's Law, which states that size is reduced by half every 18 months. However, as the components became smaller, DRAM began to gradually deviate from Moore's Law, and is now hard to shrink any further.
 
Lai believes that when Moore's Law becomes inoperative, this signals the advent of the era of spintronics. This was the insight that led to his cooperation with Lin.
 
Prof. Lin explained that in addition to carrying an electric charge, electrons spin, thereby generating a tiny magnetic pattern, similar to the millions of tiny magnets on a wafer, with the memory of 0 and 1 being determined by direction of the magnets’ positive pole. Thus it is not necessary to supply power when no operation is being carried out, and even if there is a power outage, the data will not disappear. It is expected that the use of MRAM can at least double the standby time of mobile phones and tablets.
 
Lin said that the structure of magnetic memory is like a sandwich. The upper layer is a freely flipping magnet, which can process the data quickly; the bottom layer is a fixed magnet which is responsible for storing the data; and these two layers are separated by an oxide layer. When the magnetic moments of the two magnets are in the same direction, the spinning electrons easily pass through, forming a low resistance represented as 1; when the magnetic moments are opposite there is high resistance, represented as 0.
 
However, these small magnets become unstable at room temperature due to thermal energy. Scientists have found a way to add an antiferromagnetic layer which fixes the direction of the magnetic moment magnet, but this solution also interferes with the movement necessary for the wafer to carry out its functions.
 
A major breakthrough in electron spin current
 
To solve this problem, Lai and Lin formed an interdisciplinary team with doctoral students Lin Bohong and Yang Boyuan. After a long series of experiments, they solved the problem by adding a layer of platinum, which is only a few nanometers thick and uses the coupling effect of the spin-orbit to generate a spin current, allowing for quick and accurate movement while the magnet remains fixed in place.
 
“It’s like putting a little spin on a bowling ball,” explained Lai.But if it’s that simple, why hasn’t anyone thought of it before? Indeed, the team had considerable difficulty in convincing the international scientific community of the validity of their results, and it was only after providing ironclad evidence that Nature Materials finally accepted their groundbreaking paper in the field of spintronics.
 
Opposite yet complimentary
 
In recent years NTHU has been promoting cross-disciplinary cooperation, such as the research conducted by the physicist Lin and the materials expert Lai, who have learned to work together seamlessly.
 
Their research often begins with Lin putting forward a possible solution to some problem from the point of view of physics, and then asking Lai if his solution is feasible from the perspective of materials science. This approach has also been adopted by their graduate students, who have set up a communication platform for finding innovative solutions to key problems in science and engineering.
 
However, in terms of personality, Lin and Lai are polar opposites. Lin says that he is good at coming up with new ideas, but not at bringing them to realization, but,thanks to Lai’s rigorous attitude and attention to detail that their research has borne fruit. Lai describes Lin as a brilliant thinker with an uncanny ability to make physics amazingly interesting.
 
Lai and Lin have been working together for ten years. They often get together on holidays, and occasionally even use Line to share the flashes of insight that sometimes occur in the middle of the night. Thus Lin jokes that both of their wives are starting to get jealous.
 
Since they began their joint research in 2009 their work has become increasingly influential. The impact factor of their first joint research article was a mere 3.8, while that of their second and third articles was 7.2 and 21.9, respectively.
 
Lin Bohong, a doctoral student of the Department of Materials Science and Engineering, is the first author of this paper. He said that being on a research team led by two professors who are very different from one another has been a highly valuable experience, and that he is especially thankful to his advisor Prof. Lai for teaching him how to do solid research and also for sending him to France, Ireland, Singapore and other places to participate in seminars to expand his horizons through international exchanges. Lai has also arranged for Lin Bohong to spend a year abroad conducting magnetic research.
 
In recent years a number of major international companies have begun to conduct research in MRAM development, including Samsung, Intel, and TSMC. It’s likely that mass production of high-density MRAM will begin sometime this year, a development in which the research team led by Lin and Lai has played a key role.
 
The research team is presently working on applying their groundbreaking discovery to the nano-membrane layer of other structures, and their ongoing findings are expected to have a major influence on the development of the domestic memory industry. Moreover, with the increasing prevalence of artificial intelligence, wearable devices, and the Internet of Things, the market for magnetic memory will grow rapidly. In Lai’s view, the development of MRAM technology is going to have a decisive influence on the future competitiveness of Taiwan's semiconductor industry.
 
Professor Lai Chih-huang (left) of the Department of Materials Science and Engineering and Professor Lin Hsiu-hau of the Department of Physics using hand gestures to represent the “0-1” concept in digital memory.

Professor Lai Chih-huang (left) of the Department of Materials Science and Engineering and Professor Lin Hsiu-hau of the Department of Physics using hand gestures to represent the “0-1” concept in digital memory.

Lai (center) and Lin (right) with team member Lin Bohong, a doctoral student of the Department of Materials Science and Engineering.

Lai (center) and Lin (right) with team member Lin Bohong, a doctoral student of the Department of Materials Science and Engineering.

Although very different in terms of personality, Lai (left) and Lin work together seamlessly.

Although very different in terms of personality, Lai (left) and Lin work together seamlessly.

Lin Bohong, a doctoral student of the Department of Materials Science and Engineering, is the first author of the team’s latest research paper.

Lin Bohong, a doctoral student of the Department of Materials Science and Engineering, is the first author of the team’s latest research paper.

 

No. of visitors