Simplified Spintronics

James Carman

Emerging Technologies

Spring 2003

Introduction:

In this ever changing world of technology, products of technology have evolved at an extremely quick rate. Computers have been doubling in speed in no time flat, audio recording has evolved, though more slowly, from albums now to mpeg players that can store I think around 700 minutes worth of music, and the gaming console business, not to mention all the other areas of advancement. But, getting back to computer speed and storage, the advancement in this category is starting to slow down (there can only be so many “switches/transistors” placed on a chip). So to deal with storage and speed advancements some new research is being done. One of these areas is that of spintronics. This technology is hard for me to comprehend so during this paper I will try to simplify spintronics and get down to the gist of it.

Brief History:

But before I move on I would like to give this very brief history over the subject. Moore’s law on how the speed of computers will keep on doubling is slowing down, and it has been estimated that in the next 10 years or so we will have reached the limit on speed of traditional microprocessors. So back in the early 90’s research was being done on how to make microprocessors better and faster. Research into areas like nanotubes was done, but not until 1998 could I find anything concerning spintronics and this article was talking about how Yale was doing research in the field of spintronics and how this research could lead to “quantum computers” (Moon par. 2). And from there the research has taken off with very few results but it is still promising. The one visible result from this research is GMR(Giant Magnoresistive) and this allowed for more storage on hard drives, especially useful for labtops.

What is Spintronics?:

The formal definition of spintronics is “the study of the role played by electron (and more generally nuclear) spin in solid state physics, and possible devices that specifically exploit spin properties instead of or in addition to charge degrees of freedom” (Introduction to Spintronics par 1). A simpler definition is that spintronics is a “new branch of electronics in which electron spin, in addition to charge, is manipulated to yield a desired outcome” (Fabian par 1). So how does spintronics work? To understand how spintronics work it must first be understood how an electron works, specifically the spin of the electron. The spin of the electron has three states to it; up, down, and inbetween. This inbetween state is the most important one. In today’s world of computers the spin is ignored and you have either on/off, 0/1, or up/down but with the spin of the electron you can have many states and not be limited with those two states. Because of these many states information can be processed a whole lot faster if an electrons spin carries data (Moon par 4,12). This graphic from the magazine Scientific America, June 2002, gives a good explanation on the spin of electrons:

Picture number 6 is what spintronics is all about; controlling the current flow is very important.

What are some products in research?:

The first product being developed that I want to write about is MRAM, which is also slated to come out in late 2003 or a year later. In the near future MRAM (Magnetic Random Access Memory) will replace DRAM and Flash Memory (Bonsor 1). So what does it do? For one it will enable you to turn on your computer like a TV with no boot up time. Another is that it will enable more storage and quicker access to data. The last thing that makes MRAM great is that it uses less power. It uses less electricity because MRAM takes a small amount of electricity to switch the polarity of each memory cell on a chip, compared to the constant energy supply needed to maintain an “one” within a memory cell on a normal computer. What MRAM will do can be stated in this one sentence: “MRAM promises to combine the high speed of static RAM (SRAM), the storage capacity of DRAM and the non-volatility of Flash memory.” (Bonsor 2). This picture is an example of MRAM architecture.

The next area being developed right now is the use of spintronics in semiconductors. This area would allow for “ultrafast switches and fully programmable all-spintronics microprocessors. This avenue of research may lead to a new class of multifunctional electronics that combine logic, storage and communications on a single chip.” (Spintronics 1). There are a couple of questions still needing to be answered here. One of those problems that has been partially answered is finding the right material to make ferromagnetic semiconductors at room temperature or higher. One type of material that is showing some success here are plastic semiconductors. In recent research done by Epstein and Joel S. Miller of Utah University plastic has displayed the ability to “… make all of the components that go into spintronics from plastics,’ Epstein continued. ‘So it is timely to bring all these components together to make plastic spintronics.’” (Plastics par. 8). The problem with trying to use any other type of semiconductor material is that it cannot stay magnetized at room temperature or higher. Another area being worked on with plastic spintronics is that of trying to pass a spinning electron from one computer component to another, like from one semiconductor to another. That is part of another problem in spintronics; the electron getting knocked off-kilter in these transfers and thus messing up the memory of that cell (Plastics par. 14). One other neat little fact about plastic electronics is that it also deals in the area of OLED’s which has been covered earlier in the course.

Something that is even further away and was not mentioned much in the papers I read, but could be achieved through use of spintronics is Quantum computing. I won’t delve much into this topic, only enough to say that what would happen here is that the “quantum spin states of individual electrons” would be manipulated and used to make quantum logic gates, and this in turn would enable the construction of quantum computers. Basically some of the stuff that quantum computer would be able to do is this:

Cryptography: perfectly secure communication.

Searching, especially algorithmic searching (Grover's algorithm).

Factorizing large numbers very rapidly (Shor's algorithm).

Simulating quantum-mechanical systems efficiently.

(Center par. 5).

Conclusion:

Sometime in the near future I hope our quest for smaller and smaller, and faster and faster devices will be found in the new realm of spintronics, and we will see the effects that spintronics has on the realm of electronic devices, but for now this stuff is still in the lab. So when Moore’s law of transistor placement goes down the drain Silicon Valley will be replaced by this stuff, and it might even be plastics that lead the way, (“I wonder how inexpensive plastic would make this technology and if it would be reflected in its price?”). For now I would like to end with an interesting fact on where some of our tax money has gone: “Its potential is sufficiently great that the US Department of Defense has invested more than $50 million dollars in spintronics research in the past year.” that year was 1998 (Quinion par. 1). I wish I had $50 million dollars to spend L.

References:

http://www.research.ibm.com/research/gmr.html

http://www.howstuffworks.com/mram.htm

http://www.physics.umd.edu/rgroups/spin/intro.html

http://www.qubit.org/

http://www.sciam.com/article.cfm?

http://www.glue.umd.edu/~jfabian/spintronics.html

http://www.quinion.com/words/turnsofphrase/tp-spi1.htm

http://www.acs.ohio-state.edu/researchnews/archive/spintron.htm

http://www.yaledailynews.com/article.asp?AID=4953