Quantum Experiment Shows Effect Before Cause 465
steveb3210 writes "Physicists have demonstrated that making a decision about whether or not to entangle two photons can be made after you've already measured the states of the photons."
Here's the article's description of the experiment: 'Two independent sources (labeled I and II) produce pairs of photons such that their polarization states are entangled. One photon from I goes to Alice, while one photon from II is sent to Bob. The second photon from each source goes to Victor. Alice and Bob independently perform polarization measurements; no communication passes between them during the experiment—they set the orientation of their polarization filters without knowing what the other is doing. At some time after Alice and Bob perform their measurements, Victor makes a choice (the "delayed choice" in the name). He either allows his two photons from I and II to travel on without doing anything, or he combines them so that their polarization states are entangled. A final measurement determines the polarization state of those two photons. ... Ma et al. found to a high degree of confidence that when Victor selected entanglement, Alice and Bob found correlated photon polarizations. This didn't happen when Victor left the photons alone.'
Asimov predicted this years ago... (Score:5, Informative)
http://en.wikipedia.org/wiki/Thiotimoline [wikipedia.org]
Re:Time delay - info from the future? (Score:5, Informative)
The summary doesn't say what the time delay is between when Alice and bob measure their polarization and when victor makes his choice.
FTFA:
Due to the 104-meter fiber-optic cable, Victor's measurements occurred at least 14 billionths of a second after those of Alice and Bob
Let's violate causality! (Score:5, Informative)
Or at least try...
So a key part of the experiment was that the pair of photons sent to "Victor" went through a 104 meter cable to ensure that whatever Victor did, Alice and Bob measured their polarizations first.
Presumably, one could extend this cable to increase the amount of time between Alice and Bob's measurement and Victor's decision to entangle or not.
Presumably long enough for Alice and Bob to send the result of their measurement to Victor.
And then instead of an RNG, Victor chooses to entangle based on whatever would contradict Alice and Bob's measurement.
Come on, we have to try...
P.S. the paper says they aren't violating causality, and it only looks like they are if you're looking at it wrong.
Re:Time delay - info from the future? (Score:5, Informative)
First of all, quantum effects like this don't allow the passage of information (no quantum entanglement effect does, it would violate relativity). Alice and Bob don't know if their photons are entangled simply by examining them. As a rule, quantum effects are worthless for transmitting information of any kind: both parties know what the other's state is if they know the photon's were entangled, but that is insufficient to transmit any kind of information (it is very useful for encrypting information, but not transmitting it), so you cannot build a useful transistor system using this.
Secondly, the Ars article rightly points out that concluding that effect proceeded cause should be rejected without much much better evidence. I can't explain the results, but throwing out causality so rapidly would be foolish.
One thought I had was that the detectors might actually be in a quantum state (basically, entangled with the photon they observe) after making their observation, which isn't collapsed into an entangled (or not) state with the other photon until Victor makes his decision. In other words, these results might not show up if you increase the timescale, because the quantum state of the detectors after they sense the photons (which, if it lasts long enough, can be affected by Victor after they detect the photon polarization without violating causality) might collapse before he decides to entangle the photons or not. I am, of course, not a quantum physicist, so that might not be possible.
Re:Time delay - info from the future? (Score:5, Informative)
Article says! It's on the order of 14billionths of a second.
When you say it like that, it sounds small, but if I did my math right, 14billionths of a second is the same amount of time as 28 clock cycles on a 2GHz processor.
Re:Time delay - info from the future? (Score:5, Informative)
I have a pretty bad grasp/understanding of this stuff, but if two atoms are entangled, changing the state in one affects the other, right?
No. All that happens is that when the particles are entangled they will have a correlated state when measured. e.g. if one has positive spin the other will have negative. Measuring -- or changing -- the state breaks the entanglement, so you can't simply use it like an FTL telegraph.
Besides, they are working on this now, so it hardly seems futile?
They are not working on FTL communication. The "quantum communication" they are talking about is like the GP said, in a sense a new form of encryption. You can't use entanglement to communicate FTL. However you can use it to determine if your communications have been intercepted -- due to the property that measuring the entangled particles breaks the entanglement. This is awesome because it means you could transmit a shared encryption key, and detect if anyone snooped it, and either send a new one if it was, or use the shared key if it wasn't.
Re:Time delay - info from the future? (Score:5, Informative)
Not exactly. Let me explain: when you observe a property of one of an entangled pair of objects, you automatically know the state of the other. This isn't exactly a problem, until you add Heisenbergs uncertainty principle, which states that the more you know about one property of an object, the less you can know about another (position and velocity of an electron being the classic example, but for entangled objects a better example is spin and velocity).
If observing the spin of one entangled electron lets you know the spin of both (but changes the speed only of the first, since you only observed that electron), then you logically should be able to observe the speed of the other entangled electron (which would alter it's spin... but you already know that) and know both spin and speed of both electrons precisely. This violates the uncertainty principle, so instead what happens is observing the spin of the first electron causes both electrons to change in speed, but they do so randomly: in other words, you can change one of an entangled pair by observing the other, but you cannot do so in a controlled fashion. Again, to do otherwise would be to allow one to know both spin and speed of the electron, which is impossible.
Similar logic holds true for entangled photons: observing one changes the other, but not in a controlled fashion. However, both parties can know the polarity of the other's photon (if they are entangled) just fine, which allows them to share certain secret information, which is why quantum networks are theoretically 100% secure. Anyone trying to eavesdrop will actually change the state of the photons by doing so, which can be detected. The details are, obviously, somewhat complex.
Re:Time delay - info from the future? (Score:4, Informative)
The speed of light was not the problem. The problem was the timing of the detection of the neutrino. Slight - but significant - difference.
Re:Time delay - info from the future? (Score:5, Informative)
Assuming you know enough information to determine that a particle has been disentangled (and I think that this is the case), then you have faster-than-light transmission of information.
Nope. The only way you'd know that the particles had been disentangled is when the person on Mars sent you, via normal communication channels, the information they had measured and you saw that it was not correlated with what you had measured.
That's what was going on in this experiment -- Alice and Bob could not tell just by looking at their individual particles whether or not they were entangled. Even comparing their measurements doesn't tell them, since they could have gotten the same results as they would have in the case of entanglement through chance alone. Only when Victor told them which particles were entangled could they sort their data sets into entangled and non- and see that in fact the entangled set showed the expected correlation.
BTW, this is at a high level how Quantum Encryption works -- along with regular data, you send information about your entangled particle. If the information was snooped, then the entanglement is broken, and what you measure will have no correlation with the measurements you were sent. That's the only way to tell. You can't just look at the particle and say "yep, it's entangled".
Re:Quantum Physics @ Home (Score:5, Informative)
I am an OSGS (Optical Sciences Graduate Student) and you don't need Quantum Mechanics to explain the experiment above, all you need is classical wave optics.
Linear polarization is electric field in a specified direction, lets say you have the electric field oscillating in the x direction and in the y direction for the first slit and the second slit respectively. Those directions are orthogonal to one another, so cannot interfere (the inner product is zero). But, if you have some component from both slits in some direction (for your example you will be getting out sqrt[2]/2 of the x component in the 45 degree direction and sqrt[2]/2 of the y component in the 45 degree direction when you insert the 45 degree polarizer, which is basically equivalent to the no polarizer case except you have reduced the amplitude). Then you have slit interference in the classical sense as illustrated here : http://astro1.panet.utoledo.edu/~lsa/_color/14_interference.htm, you will have to scroll down to see the two slit interference. Note that we see a sinusoidal pattern because our eyes view the time averaged irradiance (intensity) of the wave pattern, the the wave pattern itself.
What is different about the quantum case is that you can send, say electrons, through the slits *indivdually*, one at a time and they somehow interfere, that is what is intuitively strange.
Re:Time delay - info from the future? (Score:4, Informative)
Since the meter is defined as the distance light travels in 1/299792458 second
You forgot "in a vacuum."
Re:Asimov predicted this years ago... (Score:4, Informative)