Slashdot Mirror
Quantum-Tunneling Electrons Could Make Semiconductors Obsolete
Nerval's Lobster writes "The powerful, reliable combination of transistors and semiconductors in computer processors could give way to systems built on the way electrons misbehave, all of it contained in circuits that warp even the most basic rules of physics. Rather than relying on a predictable flow of electrons that appear to know whether they are particles or waves, the new approach depends on quantum tunneling, in which electrons given the right incentive can travel faster than light, appear to arrive at a new location before having left the old one, and pass straight through barriers that should be able to hold them back. Quantum tunneling is one of a series of quantum-mechanics-related techniques being developed as possible replacements for transistors embedded in semiconducting materials such as silicon. Unlike traditional transistors, circuits built by creating pathways for electrons to travel across a bed of nanotubes are not limited by any size restriction relevant to current manufacturing methods, require far less power than even the tiniest transistors, and do not give off heat or leak electricity as waste products, according to Yoke Khin Yap of Michigan Technological University, lead author of a paper describing the technique, which was published in the journal Advanced Materials last week."
You mean the 1950s are back? Tunnel diodes were supposed to rule the world back then too! How exciting!
Usually Nerval's Lobster promotes self-described tech-writer-gun-for-hire/Slashdot "editor" Nick Kolakowski's work. In this case, the author of the Slashdot content is Kevin Fogarty, who recently brought us such gems as thinly-disguised press releases for Cumulus Networks, Enterasys, and Heavy Reading, all of which use curiously-similar ambiguous stock photos from Shutterstock... My guess: the people behind the article (which we can't read) paid for it to be summarized and posted on Slashdot so they could pursue further funding by claiming their work has been "featured" (legitimized) on Slashdot.
This has been going on for some time now with Nerval's Lobster. Many people have learned not to feed the troll (don't post comments on Nerval's Lobster submissions), but if you're just joining us, welcome! And try not to feed the troll.
The reason why this is not a problem is because the electron does not actually travel from one point to the other, which would mean there were intermediary points of existence along the way. This is a quantum movement. The electron stops probably being at one place and becomes more probably in another place. It never was in any “place” to start with since placeness is not a quality of an lepton in motion.. Nevertheless, the event of the movement from one probability to the next is not really time measurable as an event, only as a measured effect.
But what if you have a mile long pole and correlate it's movements into a form of communication. As soon as you move the pole on one end it would instantaneously move on the other for instantaneous communication.
Nope. The motion propagates to the far end at the speed of sound in the pole - much faster than sound in air, but glacial compared to light in vacuum.
Don't bother looking for an unobtanium with near-infinite stiffness and an internal speed of sound faster than light-in-vacuum. The motion at one end encodes information about what is happening at that end and that information is propagated down the pole by interactions between the pole's component particles, interactions that all are no faster than the speed of light.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
When I was young kid, in the early 1960's, I visited a ham radio operator a bunch of times. Cool radios, etc. He taught me some key things about tubes, started a long slide into technology that still hasn't stopped. I asked him about transistors. He looked at me somewhat askance and said "yeah, "I heard about them things. Tubes, son. I know tubes." And went back to teaching me about tubes, and resonance, and etc. Outside of his place, I hooked into an NRI electronics course, and spent a summer sucking that down, while running to my older friend Tony to help me with the math. NRI was teaching tubes then too, but they had an excellent section on transistors, and so I grew comfortable with them just as they were becoming interesting and more widely used. Tubes, except for certain specific jobs, just aren't used much now as we all know, and I've always been grateful for my luck in terms of timing; a few years earlier, and I'd have been looking askance at transistors myself. But instead, I've been comfortable with semiconductors right up until they got too small for me to handle (surface mount, trembling hands, etc.) And I know tubes.
The idea that another revolution of similar importance may happen in my lifetime...
Damn. I just feel like one amazingly lucky fellow. :) Now, will I be able to grasp the tech if it makes it to market? That, as they say, remains to be seen. Getting older doesn't mean you're without a clue. It just means you no longer always know where you put them.
I've fallen off your lawn, and I can't get up.
Wrong. Dirac did. And it led him to the prediction of antimatter. And all quantum field theories are relativistic quantum mechanics.
What hasn't yet been managed is to unite general relativity and quantum mechanics. However that's irrelevant for the FTL tunneling question because the speed of light limit is a property of special relativity, which has completely and fruitfully been united with quantum mechanics a long time ago.
Relativity tells us that if we could travel faster than light, then we could also violate causality. So unless you are willing to give up causality, relativity indeed does ban travelling faster than light.
Faster than light expansion is just a mathematical oddity caused by the use of a particular coordinate system. Not that there's anything wrong with it, just that nobody ever seems to explain it properly.
Local distances and times are easy to measure objectively using clocks and measuring rods, but the definition is not so clear and unambiguous anymore when you're talking about large distances in the expanding universe. Different metrics exist, defining distances, times and speeds in a different way, yielding wildly different values while giving the same tangible results for any actual event. I will give you two ways of looking at the universe: the first conserves the speed of light but looks very weird, while the other looks more normal but does not respect Special Relativity. Under General Relativity, which allows a wider range of metrics, both models are perfectly valid and consistent. I will disregard the effects of gravity, but otherwise it should be a pretty accurate description, certainly enough to explain what "space itself" really means.
If you define distances, times and speeds using the common sense definitions from Special Relativity (using beams of light to measure distances, always assuming a constant speed of light), distant galaxies are traveling away from us at high speeds (but less than the speed of light) and therefore time passes more slowly for them. Since this has been the case ever since the big bang, they are younger than us at this point in time. They don't just look younger because we had to wait for their light to get here, but they really are younger "right now" even if we take the traveling time of light into account. If we could "look" at them directly without having to wait for the light to get here, like we could do in a mathematical model, we would "see" the universe getting younger and younger, and clocks ticking ever more slowly, the further out we "look" in our expanding universe. At a distance of c times the age of the universe, the big bang is happening "right now". This gives the universe a finite size (assuming nothing existed before the big bang) but it does contain an infinite amount of matter thanks to Lorentz contraction. Everything near the boundary is squished in the direction of the expansion so that an infinite amount of stuff fits in this finite amount of space.
This metric is a bit cumbersome because it gives us a special position at the center of the universe while in fact there's nothing special about our position at all. Some other, distant civilisation (in the distant future according to the above metric) will actually say that we don't exist yet and our galaxy is much younger than theirs, "now". (Using their definition of "now"). That's just the classic twin paradox, nothing really wrong with that, but it does make our point of view a bit subjective.
So cosmologists came up with a better metric, the cosmological model: they define time as whatever is measured by local clocks that are traveling at the same speed as the average galaxy in that area (the expansion speed vector), undoing time dilation due to the expansion and thereby making the whole universe the same age. Local distances are defined in such a way that objects look pretty much the same size everywhere (no Lorentz contraction due to expansion speed), which can be achieved by defining distances in function of a constant speed of light relative to the expanding universe. So in effect we stretched the universe and sped it up, just by using a different definition of "now" and by measuring distances differently. With this model, the universe looks nicely homogenous and truly infinite, making many calculations a lot easier. There's no longer anything special about our location.
But because we changed our definitions of space and time, some of the old assumptions from special relativity are no longer valid. Things can and do fly away from us at speeds well in excess of the speed of light simply because we are defining their speed differently. But the light from those places will ne