Slashdot Mirror
Stopping Light
Jon Abbott writes "NASA is reporting that physicists at Harvard University have managed to stop light altogether. The implications of this discovery are rather staggering -- quantum encryption and quantum computers might be just around the corner! " Well, I don't think this will mean any immediate changes - but it is a significant step.
And here's the story from when it was news, last year.
0xB
you just can't stop energy.
True, but you can contain it. Think of lead-acid batteries, flywheels, or any other energy storage device. The energy becomes somewhat of a potential (although in flywheels the energy is rotational), and can later be turned into useful energy.
Man is born free; and everywhere he is in chains.
The energy of the light is stored in the spinning atoms. This energy causes them to change from a base quantum state to an excited state. When they shine the second laser on the atoms, the energy is released in the form of the original laser.
The energy itself isn't really "stopped", it's transformed into a different form. When the photons of light impact an atom, it leaves an imprint (in the form of a spin). So, each unique wavelength of light leaves a unique imprint which can then be fetched at a later date by another laser pulse (or so the article says) Hope that sheds some light on the subject. :P
There's an explanation of why stopping light could lead to advances in quantum computing here.
There's also some news about it here at Harvard's site. Also, there's an article here in Scientific American about this.
Didn't this all happen last year, too? Why is it just news now?
Jack Buck (1924-2002)
Darryl Kile (1968-2002)
Slashdot Jan. 2002
Slashdot Jan. 2001
I've noticed a couple of people wondering why this discovery important. Some other people know that it is useful for quantum computing, but they don't know how it would be useful. I'll see if I can help.
The most common way qubits are stored in quantum computers is as spin, which can be thought of as angular momentum, quantum-style. The particle usually used for this task is the electron. So, now we've got the qubit stored as spin, but how do we get the different particle's spin states to interact? If we can't get them to interact, we can't do any computation, so this is a very important question.
The most successful quantum computers (those with 7 qubits) so far use Nuclear Magnetic Resonance (NMR) to make the qubits interact. This has it's problems, and would not be appropriate for a real quantum computer. So, to make a real (ie. Desktop) QC, we need something better.
This story talks about a method of turning information stored in light (as amplitude, IIRC) into spin. This sort of translation is exactly what is needed to make quantum computers work. An example QC could use a bunch of atom's as the memory system, with all of the qubits encoded as spin on the electrons orbiting the atoms. The CPU would be a bunch of optical components (beam splitters, polarizers, mirrors, etc.) that operate kind of like transistors. And the wires would just be fiber optics. Now, this is a little simplified, because it assumes we can make atomic scale optical components, but I am confident that it will happen soon.
Hope this helps some people understand why this is Stuff that matters.
Yes, I'm still a junky. Are you still a bitch?
"Light of Other Days"
Bob Shaw
First came out in '66
Still gives me a lump in my throat just thinking about it.
You either believe in rational thought or you don't
We've known that light can slow down since 1850. The speed of light does change when traveling through various things, depending on its index of refraction. So, you are not correct -- light can be and is slowed down easily.
This is the four thousandth time this article has been posted here and it is still doesn't follow. Nobody is freezing photons, they're just getting them stuck in the middle of some molecules so they have to wait for another laser to be able to knock the photons loose again. Stopping photons is not the same as trapping them.
I'm a loner Dottie, a Rebel.
but this phase information is quickly lost as the atoms move around in a thermal equilibrium. think about it as sky-writing. the information is written there, but as the particles move around the infomatino is quickly lost.
most of these experiments have been done with UltraCold atom clouds, and the most recent ones (presented at DAMOP last year) were done in Bose-Einstein Condensates.
due to the very short "coherence time", this phenomena is most likely not very useful for quantum computing.
the buzzwords to look for when it comes to quantum computers (i.e. the things most likely to work) are "trapped ions" and "optical lattices". i promise, one of those two will be used in the first functional quantum computer.
muerte
IIRC Einstein proved that light is particles that moved like energy (in a waveform).
Sort of. It was shown that particles behave like waves and vice versa. Thus, since light is a wave it can also be thought of as a particle. We refer to this particle as a photon.
However, it's not a particle in the same way as an electron or a cookie crumb. It has no mass and, thus, exists only in the form of energy.
This was published in Nature over a year ago (25 January 2001 to be precise). This article (PDF format) is a nonspecialist introduction to this work, and this article (PDF format) is the peer-reviewed research article from Nature.
"It take 9 months to bear a child, no matter how many women you assign to the job."
Assume you make an incredibly good mirror: it's 99.999999% reflective. (How you're going to manage to do this while still pumping light in from the outside is unclear- 1/2 silvered mirrors are exactly that.) No mirror is even close to this value, BTW- the best around can do about 99.99% or so.
Assume you have a 1 m diameter ball. Light travels 300,000 km/sec: 3e8 m/s. Thus, you get 3e8 collisions with the mirror every second. Total saved light= 0.99999999^3e8 ~= 0.05. In other words, after 1 second only 5% of the light remains.
"Photon torpedoes" supposedly use matter-antimatter as a power source: pure mass-> energy conversion- why bother with light at all?
Eric
"Seven Deadly Sins? I thought it was a to-do list!"
I know you were joking, but you really can do Quantum computing with Perl
-- Will quantum computers run imaginary-time operating systems?
The article mentions experiment from 1999 where light was slowed to the "bicycle speed". This was accomplished by shining light trough "Bose-Einstein" condensate. Bosons are particles with integer spin (e.g. photons). In 1924 it was predicted that an ensemble of bosons could, under certain conditions, undergo a phase transition. This is analogue to vapor condensation or crystallization of liquid. In order to create Bose-Einstein condensate it is necessary to achieve temperatures less than one millionth of a degree above absolute zero. First successful experiment was performed in 1995 utilizing laser cooling. One of the properties of BE condensate is that the light propagates trough it with speed that is 20-million-fold slower than a speed of light in vacuum.
h tm l,
The article is not very informative about actual physics involved in the newest experiment. However there is a nice description at: http://www.aip.org/physnews/update/521-1.html.
Also there is a an interesting site about Bose-Einstain condensation at
http://www.colorado.edu/physics/2000/bec/index.
with some nifty Java applets
It's "its", not "it's"!!!!!!!!!!!!!11
-joe