Slashdot Mirror
More on Spintronics
segment writes "'We have discovered the equivalent of a new 'Ohm's Law' for spintronics - the emerging science of manipulating the spin of electrons for useful purposes,' says Shoucheng Zhang, a physics professor at Stanford. 'Unlike the Ohm's Law for electronics, the new 'Ohm's Law' that we've discovered says that the spin of the electron can be transported without any loss of energy, or dissipation. Furthermore, this effect occurs at room temperature in materials already widely used in the semiconductor industry, such as gallium arsenide.'"
The applied electric force, the spins and the spin current align in three different directions that are all perpendicular to each other ''This is a remarkable thing,'' explains Zhang. ''I push you forward and you move sideways - not in the direction that I'm pushing you.''
Same thing happens with me after about a six pack.
In maybe a 10-year timeframe, spintronics will be on par with electronics
If the actually manage to go from idea to commerically competitive "spintronic" circuitry in only a decace, I'll consider that proof of some sort of space-alien technology transfer deal going on.
Gentlemen! You can't fight in here, this is the War Room!
I had a prof in college who loved to tell the tale of finishing his bsee at mit, and didn't know what to do next. Being young and stupid (his words) he went down to the army recruiting office and inquired about electronics. The guy behind the desk says he has to come back the next week to take a test.
So, he shows up with a bunch of other hopefuls (again, his words), and takes the test. One of the questions is 'state the 3 forms of Ohm's law." As a good ee, he immediately writes down 'V = IR', no problem. Thinking (!!), he remembers there is a form involving current density, and sets about to derive it (in class this included the steps he took). Now, a third form. He drew a blank, so went and finished other parts of the test.
Coming back to this question, he's suddenly inspired by remembering something based on magnetic density in a coil or transformer. Again, he sets out to derive the equation, but the guy giving the test says 'times up' before he can finish.
The next day, he heads back to the recruiting office, and asks how he did. "Great," says the sargeant, "but, what was up with the Ohm's law question?"
"Oh, the standard form, and then experessed in current density, and...."
"Nah, all we wanted was V = IR, I = V/R, and R = V/I."
Proof of, once again, that engineers, like musicians should not try to be funny.
At subatomic levels, every process is 100% efficient. The basic principles that you learn in mechanics which warn you that there is no such thing as a perpetual motion machine, etc... are results of statistics and macroscopic effects. Microscopic is not miniaturized macroscopic.
effectively allowing communication based upon movement.
The rod would move at the speed of sound through its medium (the speed of sound varies largely depending on its medium). IN any case, it would be MUCH slower than the speed of light.
Currently, the only thing confirmed to move faster than the speed of light (confirmed via the "alan aspect" experiments, if you want to google it), is the spin on a pair of electrons. Two elextrons in a pair alwats spin in reverse directions. Even if the two electrons are 1000 miles apart, if you polarize one (change the spin), then the other spin will reverse itself instaneously.
This was tested by alan aspect (who built upon the EPR thought experiment), who subjected two electrons traveling in opposite directions to a polarizer and found that the correspondency between the two electrons meant that there HAD to be some osrt of faster than light communication (it violated "bells theorum" if you want to do more googling). That is, it wasnt a coincidence, or due to 'hidden variables' as einstien thought. It truly was faster than light communication, somehow, between the electrons.
This is the main discrepancy between Einstein's relativity and Bohr's quanutm theory (Einstein's theories actually pushed quantum theory, ironically). Einstien's relativity theory states that should anything move faster than the speed of light in the spatial dimensions (x,y,z), it must move backwards in the fourth dimension (time). Basically, he argues that everything moves through the four dimensions (x,y,z,t) at the speed of light. Photons move through the spatial dimensions (x,y,z) at the speed of light, and thus do not mvoe through time at all. The photons that exist now have not aged at all since the big bang. This is how einstien explains "Time dilation." This has been confirmed a number of ways, most easily by clocks on airplanes. Clocks put on airplanes, which move through the spatial dimensions(x,y,z) through high speeds (high being relative to normal human movement) have been found to register less time than their "at-rest" counterparts. Of course, quantum theory somehow defies this concept. String theory explains this by offering multiple dimensions past the 4th (I beleive steven hawking's count is at 14 right now)...
Not so sure about quantum computers, but i belive this is the idea behind them. Transistors used now read either High or low, +5v or 0v, which correspond to binary terms of 0 or 1. Thus we can gather data by reading the charges on the transistors. If we could use electrons, a up-spin meaning 0 and a down-spin 1 (not really up or down, but thats how we denote them), then we could use a 100% efficient replacement for transistors.
If anyone wants to correct me, please do. I havent taken a physics course in my life (yet) and am probably wrong about some (most) of what i just said.
the byproduct of years of oppression by the white man
If I recall correctly there are a few problems with this method of transmission:
... its a bit more complicated than that.)
1. Once you transmit using a particular electron pair you can't use that pair again, so you have to pre-prepair as many electron pairs as you think you will need for a transmission.
2. Creating perfectly isolated pairs is difficult. The basic problem is making sure the pairs you create aren't entangled with any other qubits (and using extra bits to do error correction because its next to impossible produce pure states) People are working on efficient ways to do this, but although it won't be prohibitive for, say, prearranged data transmission it really wouldn't be economical for circuits.
This article is talking about something else aparently: some kind of wave of spin -- like a current.
(BTW in the method you're talking about one doesn't exactly "change the spin"
your reffering to the Alan Aspect experiment, which was built on top of the EPR thought experiment(Einstein-podolsky-rosen.
the byproduct of years of oppression by the white man
Spintronics is promising, but I doubt that it will be the NBT. Quantum-dot Cellular Automata (QCA), which encodes binary information based on electron orientation, seems to hold more promise. It is highly scalable, small, can hybridize with CMOS, and can already be fabricated at low temperatures. With the addition of clocking regions to lower inter-dot tunneling barriers, even pseudo-pipelining is realizable. Perhaps the best thing about this is that it all cells are coplanar! I just attended a conference (IWQDQC) on Quantum Computing, and believe me, spintronics faces its share of problems.
Ohm's law describes the creation of a current by the application of a voltege. This new law seems to describe the creation of a, um, whatever you want to call the "movement" of the spin of an electron, by the application of an electric field. Or more accuratly, it probably describes the movement of a group of spins.
They're describing generally the same kind of action, at least viewed in a certain way, in two different kind of "substances."
This Space Intentionally Left Blank
If you read the abstract for the actual paper you'll
see that they are basically talking about a more
sophisticated version of a quantum-hall effect,
i.e. they are talking about the evolution of a
correlated state, a different one from superconducting
condensate or bose condensate but another type
of correlated state. Correlated states can result
in negligible dissipation (e.g. superconductivity
or superfluidity). They will not be immune from
thermal fluctuations esp. at room temperature nor
will they be immune from dissipation at impurities
and such. But other than that having spin supercurrent
seems quite possible.
And I am a graduate student doing physics research
in the are of high-temperature superconductivity.
Mr. Zhang is quite well known in this area since
he proposed a so called SO5 theory which aimed to
explain everything about high-Tc in one elegant
formalism (his theory is oversimplified at best).
He has worked with Bob Laughlin a lot lately (Laughlin
got a Nobel prize for his theoretical work on, you
guessed it, quantum-hall effect). So these people
are legit, they know what they are talking about
but Zhang has been known to throw wild ideas out
there (and more often than not even those have
at least a grain of truth in them).
Sorry, no.
Entropy is not an absolute law, but a law based on extreme probabilities. In any reaction, certain quantities are completely conserved. One of these is energy.
The increase in entropy that occurs is due to energy being converted into less usable forms, such as from motion (kinetic energy) to heat (thermal energy).
It is not actually a decrease in total energy. Energy is perfectly conserved in any reaction.
In subatomic reactions, there is no place for energy to go, so to speak. In fact, the only thing energy really is is the motion (and mass, though those are remarkably interlinked) of subatomic particles.
When two subatomic particles collide, if neither of them splits or gives of any other particles, the energy remains entirely in the two particle system (that is, all that changes is kinetic energy; speed). Well, direction changes as well.
Mod parent down; he is incorrect. (or not, I'm actually in favor of the mod up only philosophy, but parent would be a good one to mod down if you believe in modding factually incorrect posts down).
For to end yet again.
Yes, but the receiver would get nothing but noise since you cannot control the direction of spin. You'll have only a probability of the spin being in a specific direction. So you cannot modulate the signal to send information. The receiver will only have corresponding results if you compare notes later. That is, sender and receiver will always have corresponding states of their entangled particles, but because the states are random they cannot carry information. It's presently a metaphysical question about how entangled particles always match states over long distances. This "quantum nonlocality" is one of the greatest scientific mysteries of the world.
The discussion on spin is wrong. Spin has nothing to do with the rotation of macroscopic objects like the Earth, it's an intrinsic quantum property of particles like the electronic with no macroscopic analog. The best explanation I've heard of spin that doesn't involve explaining the details goes like this: spin is a measurement of the number of rotations required to bring a particle back to its initial state. One-half spin particles, like the electron, require, counterintuitively, two full rotations to go back to their initial state.
The physical situation seems to have very little to do with Ohm's Law except in the loosest sense. They're describing a current consisting of electron spins under an external electric field. This has some interesting properties (I'd like to poke at the math, if I could read the paper), one of which seems to be that it is predicted to persist at much higher temperatures than the best superconductors. If so, because this spin current seems to be dissipationless, this would allow information to be transmitted without generating heat.
Interesting stuff; a pity the article was so poor.
Not all devices are linear and follow Ohm's law over wide ranges of voltages/currents. Sometimes there's an exponential relation, or others. For example, in a superconducting filament, one has bizarre quantum effects kicking in for the effectively 1-D system, and the effective Ohm's law has the voltage proportional to exp[I]. Only linear (and hence Ohmic) at small currents.
Then there's the Hall Effect where a current flowing through a wire (can be a thin foil) with a perpendicular magnetic field will cause the current carriers (either electrons or holes) to drift to one side or the other of the foil [F=q(v x B)] where the F is the force, v is the carrier velocity, and B is the magnetic field. x is a cross-product (v and B are vectors, so is F). In other words, the force acting on the carriers is perpendicular to the B-field and the current velocity, and creates a transverse voltage, often called the "Hall Voltage". So you now have a current creating a transverse voltage, which lets you apply a variant of Ohm's Law to define a Hall Resistance, sometimes called Rxy, where Rxy=Vhall/I (could be a non-linear relation too).
So in this case of spintronics, they define another variant of Ohm's Law to relate the current of the spins in relation to an applied electric field. Note that the transfer of spins across the device probably doesn't correspond to the actual transfer of electrons, but a signal propagation of spins instead.
Finally, there are other cases where one can have current flow without resistance. One case is superconductors. Another is the so-called Quantum Hall Effect. However, both of these occur at cryogenic temperatures.
make world, not war