Spintronics

[Pat Says]

"His thesis focuses on the spin dependent transport of electrons at material interfaces."

 

Sorry, this is kind of vague... It's my brother's thesis and I was wondering if someone has some kind of idea what it's about. I sort of got that it has to do with the spin-up/down of electrons. I think it may be leading up to quantum computing or something. If someone could explain or give an idea as to where they think this is going I would appreciate it very much. I would ask my brother but he is very busy and lives far away. Thanks.

[5614]

Well you haven't given much detail, but a "spin dependent transport" system would be one which, presumably, transports only a specific spin (in this case) electron. It could also be that say spin-up electrons are transported one way whereas spin-down is transported in a different direction, or not at all, or at a different speed... all we know is that the transportation or movement of the electrons is somehow effect by their spin.

[Pat Says]

So is this variable spin where computing would come into play? (instead of 1s and 0s?)

 

And how would this come into play with "material interfaces."

 

Sorry about the vagueness. I'll see if I can find out a least a little more on what he is doing... I'm interested but I don't know even know what to research.

[5614]

Quantum computing works by using the spin (e.g. so here we have an electron with either up or down spin) where either up would represent 1 and down 0 or vice versa... so we have a quantum property (spin) replacing electrical pulses like we have in a current computer (where if theres a current it's 1 and if not its 0).

 

"material interfaces" I assume this is the material where this "spin dependent transport" system operates, ie. it is on this substance (called the material interface) that the action ("spin dependent transport") actually happens.

[Pat Says]

So (sounding completely ignorant), what makes quatum computing so much more capable than normal 1's and 0's computing if it is only replacing the 1's and 0's with spin. I know..er I think I read, that it can carry out multiple operations at a time and is not necessarily faster but how does it multi-task?

[5614]

Quantum Computers (QCs) use qubits (like a bit, but the quantum version!) and so don't have to be in a set state of either 1 or 0 because they can be in a superposition (ie. we do not know what state they are in). A QC with x-qubits can then try multiple solutions (apparenlty) simultaneously. So the x number of qubits could assume [math]2^x[/math] states all at once, and each of the numbers from 1 to [math]2^x[/math] could be tried simultaneously to see if it worked.

 

This:

http://en.wikipedia.org/wiki/Timeline_of_quantum_computing

Is quite a neat timeline on QCs... also see Shor's algorithm:

http://en.wikipedia.org/wiki/Shor%27s_algorithm

QC info here:

http://en.wikipedia.org/wiki/Quantum_computer

And you from there will see what I mentioned above about superposition and also quantum entaglement (which, if you read about, may lead you to quantum teleportation... both of these subjects are found in many threads, please use the search feature on this site to find them... and NO neither are FTL (faster than light)).

[Pat Says]

Woah... I see where the potential of Quantum computers lie. I could carry around my effing pocket quantum computer and put today's fastest supercomputer to shame. Does the government not realize this? I would be putting lots of money into this and the education of it. That's incredible. Thanks for all the explanations and the links, they are great.

 

edit* and with that amount of computing power it has got to be about equal if not greater to our brains abilities... meaning AI!

[5614]

I don't know what the government knows, what I do know is that scientists are working on it at the moment.

 

A fast computer doesn't mean AI, all it means is that it can do current commands just a tad faster! Lets not get sidetracked with AI... again there's threads on this already.

[Pat Says]

Yeah, I'm just saying that something that could be that monumental should have a little more interest put into it.

[5614]

Obviously, there is a lot of interest and work going into QCs at the moment... but it's not easy work!!!

 

If you think it took about 40 years from the very first computer to get what, by today's definitions, is a standard computer [even if it was a bit slow] and then compare that to the first QC built 7 years ago, we're already on 7qubit so they're coming along.

[Pat Says]

It's not easy? Pssh, sounds like something I could build in my basement.

 

I think it should be in the news more to spark the average person's interest. If that happens then funding will increase and it will come to fruition faster.

[DQW]

... all we know is that the transportation or movement of the electrons is somehow effect by their spin.

While that last clause is true, the previous parts are not really accurate.

 

Spin dependent transport is a very active field of study in solid state physics and materials science; especially now, with the applications that spintronic devices have in building sensors, RAM chips and hard drives.

 

Pat: You probably know that electrons have a property called spin, that is responsible for the magnetic behavior of materials. When the spins of neighboring atoms interact with each other, you have ferromagnetism (FM - where all the atomic spins want to point the same way, say up) or anti-ferromagnetism (AFM - where the spins want to be alternatively up and down in a crystal).

 

Now, electrical conduction is a process where the electrons hop from one atom to another (loosely speaking; but this is truer in semiconductors and insulators than in metals). The spin of this particular (valence) electron is forced to be a certain way (either up or down), depending of the spins of all the other valence electrons in that atom. (See Hund's Rules and the Pauli Exclusion Principle). So, when the electron hops from one atom to another, the ease of hopping depends on the spins of the valence electrons in those neighboring atoms. So, for instance, it might only be possible for the electron to hop if the neighboring atom was aligned the same way (ie: the material is ferromagnetic). In other rare cases (look up Superexchange), having an AFM ordering might help with the hopping. This is essentially what spin dependent transport is about.

 

In the context of spintronics, one is particularly interested in a phenomenon called Giant Magnetoresistance (GMR), where just as described above, the resistance of an interface between two materials depends on whether the spins on either side are alined or unaligned. If the spins are lined up the same way (this can be done for a ferromagnet, by appliying a magnetic field) on both sides, the electrons find it easy to jump across, and so the interface has a low resistivity. If the spins are misaligned or anti-aligned, it's very hard for the electrons to be happy after jumping across - so they don't, making the resitance of the interface large.

 

That's briefly what spin dependent transport is about.

 

What follows now is a slight digression, based on something said earlier in this thread.

 

...spin-up electrons are transported one way whereas spin-down is transported in a different direction

If you said this (as a matter of fact) to a condensed matter physicist, you will evoke the strongest expression of surprise and bewilderment (disbelief, most likely though) that you've seen in a while. There is no existing mechanism by which the spin state (at the Fermi surface or equivalent) determines the polarity of transport. In fact, such a thing would be considered extremely bizarre - especially since no one has observed such a behavior...

 

...until fairly recently ! One American (Gramila, et al) and one German (von Klitzing, et al) group have each independently measured what is now called "negative coulomb drag" in double quantum well structures in the Quantum Hall regime. Only in this very special phenomenon does the spin state appear to affect the polarity of transport - and to date, there is no good theory that explains this bizarreness !