Interface engineering of spin-tunnel contacts to silicon:
Towards silicon-based spintronic devices
Promotion date: 31 August 2007
Roughly speaking modern electronics is based on the transport, manipulation, and storage of charge, with the transistor as the central element. But the electron not only has a charge, but also a spin. My work concentrated on the silicon-based spin MOSFET, a transistor using the magnetic spin property of the electron. The Schottky barrier has always presented a problem in developing transistors based on spin. Until this moment, no one succeeded in realizing the spin MOSFET which combines a magnetic layer with a semiconductor. We managed to tackle that problem. A gadolinium layer of no more than one nanometer in thickness combines the magnetic world with electronics. In this way, it will be possible to put a magnetic memory element directly to a silicon transistor.
Why is that magnetic element so important?
Memory directly coupled to processing power is an attractive and energy efficient option. Magnetic memories, like the hard disk, are based on technology that is totally different from the technology of electronic circuits. Combining the two is important because a magnetic memory doesn’t need additional energy to keep its content: once put into a memory state, it keeps this state. A magnetic layer, placed on a transistor, yields a powerful new component combining memory and processing power in a very direct way. This is good for cutting energy consumption, of vital importance in mobile devices.
So this is something of a breakthrough?
Yes, you could call it that. We published an article in Nature Materials about it.
But can it be applied in industry?
Now that the contact problem has been solved, others can start designing new components in which electronic and magnetic technology is combined.
Where are you from?
I am from South Korea.
There is a very important industry in South Korea, isn’t there?
Yes, in the field of electronics, no spin electronics as yet. I worked in a company called LG electronics before I came here. I had been working at superconductivity for 7, 5 years, which for the company was not viable commercially. So I decided to do a PhD in spintronics. I searched for groups all over the world and found that the group at Mesa was the best for me.
Here the principle of the spin valve transistor was developed in 1995. I send an e-mail to Ron Jansen and was invited for an interview. I was shown the facilities and was impressed.
Had you ever heard of the Netherlands before?
Actually in 2002 there was a world cup in South Korea. Guus Hiddink was the coach. But I had also links via my sister and a friend.
What did you like best in the past four years of you PhD?
The most important thing for me was the research. I liked that a lot. I also met a lot of nice people, colleagues and friends. I got excellent support, Ron Jansen’s (promoter Cock Lodder, co-promotor Ron Jansen) door was almost 24 hours open and I could discuss everything with him.
What didn’t you like?
I could not say, I don’t even have serious complaints about the weather.
What are your plans?
I have a position in South Korea at a research institute. I shall do the spintronics research.
In this I have exactly what I set out to do when I first came here.
When shall we have the first silicon-based spintronic devices?
That is a very difficult question. If I could answer that, I would certainly become a rich man. It is my guess that we need another ten years to get everything ready in terms of science and commercial development.
For the summary of the thesis, click here (pdf)