Quantum entanglement is weird. And I mean weird even for quantum mechanics, which sets a whole new level for weird.
 |
| Why is Dick York’s picture here? Is it entangled with this post? Spooky. |
In the world we live in, the macroscopic world, if you flip a coin, it can either land heads or tails (ignore it landing on its side, the results of which have been thoroughly discussed in The Twilight Zone). That’s it. But in the quantum world, that coin could be both heads and tails, with an equal probability of both. Until you actually examine the coin and see which way it landed, it exists in that weird in-between state.
I’m not making this stuff up. This is very real, as real as atomic bombs, computer chips, and solar cells. In fact, all three of those depend on quantum effects, so we know, with absolute certainty, that quantum mechanics works.
Now imagine two coins. They are rigged in such a way that if one of them is flipped, the other will invariably land the other way. So if coin A lands heads, then coin B lands tails.
Flip coin A. Don’t look yet! What state is coin B? Well, it’s either heads or tails, but which one depends on coin A, and we don’t know that yet. In fact, since we haven’t looked at coin A, it’s both heads and tails according to QM. So coin B is as well, but opposite (think of it as tails or heads).
So let’s look at coin A. Oh, it’s heads! So coin B is tails.
OK, cool. Now repeat the experiment. But this time, before looking at either coin, glue them each to a piece of wood, seal them in separate boxes, and move them hundreds of kilometers apart. Mind you, they are both in their heads-and-tails/tails-and-heads states. OK, after they have been separated, then look at coin A. Heads again! That means suddenly coin B is tails.
But wait a sec. How did coin A tell coin B which state it was in? They couldn’t talk to each other; they were too far apart!
If that bugs you, well, it should. It bugged Einstein a lot, and no offense or anything, but he was probably a lot smarter than you. But he never really liked quantum mechanics, and he dubbed this "spooky action at a distance". In principle, this could be done light years apart, where the coins are so far apart there is no way for them to communicate.
This was just a thought experiment, and in the real world coins won’t work. They’re too big for QM rules to apply. But you could do this with photons, which obey the rules of QM to the letter. There are ways to set this up, and in fact it’s been tested on Earth many times.
It works. I’m no Einstein (my hair is better, marginally), but this freaks me out. QM is really really weird.
This effect, called quantum entanglement, isn’t completely understood. There are different explanations for it, but they’re hard to test, so it’s hard to know which ones may be right and which may be wrong. The problem is, using photons limits you to line-of-sight: you can separate the two photons, but you have to be able to see both of them to measure this effect. This has been done with baselines of over 100 kilometers, but that’s not far enough to distinguish some of the hypotheses.
So some European scientists came up with the idea of using the International Space Station (I know! Using ISS for science! Wow!) to test this out. They can create a small setup with a laser which can create entangled photons. The entangled photons are then sent simultaneously to two different ground stations, widely separated on the surface of the Earth, so that both have a copy of the entangled photons. In addition, two quantum keys are created based on the photons; this is essentially a code based on the state of the photons — like winning a bet is based on which way a coin lands. The two keys are different, and one each is sent to the two ground stations. So both stations have a pair of entangled photons (identical to the other station’s) and a different key.
Each key is actually a long chain of 1s and 0s. The two keys are then compared on the ISS to create what’s called a bitwise XOR — for example, if two coins both land heads then the XOR operation yields a 0, but if they land differently (one heads and one tails) then it yields a 1 — it’s just telling you whether they are the same or different. So for each place in the key, the two numbers are compared, and if they’re the same (both 1s or both 0s) then a 0 is written down. If they are different then a 1 is put there. When this is done, you get a third string of 1s and 0s, representing a comparison of the two keys.
Still with me? Yeah, me neither, but we’re almost done. So now the ISS has this long number string which represents whether the keys are alike or different. It then transmits this to one of the two stations on Earth.
So? What does this mean? This means that now the two ground stations can create a code between them based on their keys, a code that is known only to them and no one else. Furthermore, this code cannot be cracked by anyone, anywhere, because it’s based on entangled photons that cannot be known to anyone else! Because of entanglement, they know what the other station has because they can look at their key and figure it out. But no one else can.
If I am interpreting this correctly — and no guarantees, me droogs, because after thinking about this for an hour now my brain cells are starting to turn into oatmeal — then we have a code that is basically impregnable. No one else can tap into it, making it pretty secure. And this can be done over thousands of kilometers, not hundreds like on Earth.
Now, I have readers who are more familiar than I am with this stuff, so please, if I am misinterpreting this please hash it out in the comments section. But if I understand this correctly it will have all sorts of uses in encryption, as well as helping us understand quantum mechanics. And by "us" of course I mean other people who already do understand these aspects of quantum mechanics. As for me personally, I simultaneously understand and don’t understand it.
Which means I understand it.
Hmmm, I may have just collapsed my wave function. I think maybe now it’s time for me to stop worrying about this and instead just go play with my cat.
Tip o’ the phase-shifted aggregate wave state to Larry Klaes.
June 9th, 2008 at 9:01 pm
Phil, Einstein described what Schroedinger and others proposed about entanglement as “Spooky action at a distance”, not his own hypothesis. That’s when he reached the EPR Paradox, because he believed that there was no communication between particles when others did. It’s a paradox because in QM all possible states have to be assumed for a system, so if there is no communications between the particles, they’d have to be in a predetermined state beforehand, thus directly contradicting the law.
June 9th, 2008 at 9:07 pm
I’ve always found this fascinating, and it’s interesting to see it being put to this sort of use. I only wish I could wrap my head around it enough to actually be able to explain it thoroughly to someone else well enough to sound smart.
June 9th, 2008 at 9:14 pm
So has anyone tried to measure the speed of entanglement communication? Maybe there is something that leaps between the particles informing on the other. If it truly is instantaneous, then QM is right (if I halfway understand what I think I do about it).
JBS
June 9th, 2008 at 9:14 pm
This is amazingly interesting. I´ll go find a creationist and try to explain this to him/her. Yep, I´m that cruel
June 9th, 2008 at 9:17 pm
But is your cat dead or alive?
June 9th, 2008 at 9:23 pm
Pfff. Your hair is SO not better than Einstein’s.
June 9th, 2008 at 9:25 pm
Probable typo in “so we know, which absolute certainty, that quantum mechanics works”.
Once in my life, I dropped a nickel that came to rest on its side. So one-in-a-million may be accurate odds on that.
June 9th, 2008 at 9:26 pm
It doesn’t bother me at all.
But then again, *I’m* spooky.
BTW, the Dick York gag was the funniest thing you’ve ever done.
June 9th, 2008 at 9:55 pm
Is this the start of the ansible?
June 9th, 2008 at 9:57 pm
Aha, quantum cryptography, one of my favourite subjects. Here’s my take on how this all works.
The first bit has nothing to do with quantum mechanics, and deals with the idea of a “One Time Pad”. Here’s how that works. I give you a series of random letters on a “pad”. You want to send me a message. So you write out your message one letter at a time, and write my letters underneath them (and it’s important that I gave you at least as many letters as are contained in your message – if you want to say “HELLO GAZ YOU ROCK” then I need to have sent you at least 15 letters, as you have to start from the beginning of the pad and you can’t “wrap around” if you run out). You convert each of your letters and my letters to a number – A is 1, B is 2, Z is 26 – and add them up for each letter. Then you subtract 26 from each number if it’s higher than 26 – basically you’re making sure that each of your totals is in the range 1 to 26 (dividing modulo 26, if you know what that means) – and then convert it back to a letter. Let’s say that your pad contains “AAAAABBBZZZYYYY”. You would end up with the message “IFMMP ICB YOU QPBJ”. You then send me that message. I have a copy of the pad I sent you – the ONLY copy, ideally – and I can use a similar process to decrypt your message (and thank you, yes, I do rock).
It’s important to recognise that this code is absolutely unbreakable. It doesn’t matter how powerful computers get or how much time you have – it’s completely impossible to break this code without the pad that I sent you – and we only use that pad once (hence the term “one time pad”) so that any future messages will have to use a different one time pad. The problem, however, is exchanging the pad – the key, if you like. It’s going to require a secure channel to exchange pads. Let’s say that I generate 640 Mb of letters (I’ll actually use bytes instead, so that I can send arbitrary language messages, but it’s the same principle). I can stuff that on 2 CDs – one for me, one for you – and then hand one to you. But I can’t stick your copy up on a website for you to download, because:
a) I can’t guarantee that you’re the only one that will download it.
b) You can’t guarantee that I’m the one that put it there.
Such a method is therefore vulnerable to both passive eavesdropping (someone else can download it, or “sniff” the network packets as you do) and malicious tampering (hacking the site so that you get something different to what I wanted to send you). It is thus unusable in a situation where the parties can’t securely exchange keys. Public key cryptography – which is basically what your SSL session uses on a banking web site – gets around this problem with asymmetric encryption and decryption processes; all you need to know for the sake of this point though is that while very secure and not needing me to physically hand you anything, it CAN theoretically be broken if you have something capable of factoring very large prime numbers quickly (interestingly something that a quantum computer may well be capable of).
So instead let’s look back at the one time pad idea – remembering that it cannot be broken, even by a quantum computer capable of running squillions of times faster than any modern PC, and even if we let it run for the entire length of the universe. Times two.
Let’s suppose I want to send you a message consisting of 1s and 0s (which is enough for me to send any message at all). We’ll call each 1 or 0 a “bit”, and let’s say I want to send you a message 100 bits long. If I can manage to do this with a one time pad, then if we can securely exchange a pad that is 100 random bits long, we’ll have a one time pad that can’t be broken. Is there any way for us to send this securely?
Well, yes. What we do is agree that I will send you photons that are in specific phases. For illustrative purposes, let’s say that if I send you either a “|” or a “\”, then that’s a 0. A “-” or a “/”, on the other hand, is a 1. You set up a filter on your side that randomly switches between “+” and “X”. The former will let a “-” or “|” through; the latter will let either a “/” or a “\” through.
Now here’s where it gets interesting. If I send you a “|” and you have randomly selected the “X” filter – because remember, I’m randomly sending out these bits, and you’re randomly choosing filters to receive them – then what happens? A “|” can’t fit through a “X” shaped slot, right? Well, it turns out that it WILL go through, and it will come through shaped as either a “/” or a “\” – with exactly 50% probability of each, and no way to predict beforehand.
How do we use this? I send you about 300 bits that I’ve randomly generated – randomly either a 1 or 0, and randomly selecting one of the two ways to send each one. (It should be noted in passing that I’m using a special laser to do this, traditionally through a fiberoptic cable, but it can also be done at greater distances through the air with special techniques that the ISS is apparently using to greater effect than ever before). You’re randomly choosing a filter each time, and recording whether or not it comes through a “1″ or “0″. Now, about 50% of the time, you’ll have chosen the wrong filter, so the bit that you got might be wrong. Ouch – how can this work?
No problem. I send you an email, telling you which shaped filter you should have been using for each bit. For example, if I sent you “||–/\\/” then my email would say “++++XXXX”. You compare the filters you actually used, and tell me which ones you got correctly. Note that my email didn’t say whether the bit I sent was a 1 or 0, and neither did yours – anyone who intercepted that email learns nothing of use. (And it gets better – see below). On average, you’ll have gotten about 50% of the filters correct, so we’ll end up with about 150 bits that we can use as our one time pad. My message is only 100 bits long, so that’s enough – I can send you my message now, and you’ll know that the code is unbreakable, so it’s no longer important for me to make sure that nobody can intercept the message.
But what if someone intercepts the transmission of my key to you? This is called a “man in the middle attack”; basically, some nefarious person would set themselves up as the receiver of my 300 random bits, and then retransmit them to you. You’d never know, I’d never know, it’s all useless, our one time pad has been intercepted…
… except that this won’t work. Remember that he doesn’t know what filters to use any more than you do. Suppose I send you a “/” (0) and he has a “+” filter set up. 50% of the time, he’ll get a “|” (1) and the other 50%, he’ll get “-” (0). He won’t know that he chose the “wrong” filter, and he’ll retransmit what he thinks it was – 50% of the time, he’ll retransmit the wrong bit. Remember that we ended up with 150 bits we thought we could use, and we only need 100. So before sending you the message, I randomly select 50 bits out of that 150 and tell you what they were. You compare with your list. If there was a guy in the middle making mistakes, then some of these bits will be wrong. And that means we’ll know that there’s an eavesdropper! (If they all match, we just discard those 50 bits and use the other 100 as our pad).
So quantum cryptography is not only unbreakable – totally unbreakable, as it’s a one time pad – but it’s also impossible to eavesdrop upon without being detected.
June 9th, 2008 at 9:57 pm
“Spin State” by Chris Moriarty
Read it.
http://www.amazon.com/Spin-State-Chris-Moriarty/dp/0553586246/ref=pd_bbs_sr_1?ie=UTF8&s=books&qid=1213070103&sr=1-1
June 9th, 2008 at 10:02 pm
Obligatory:
“Wait a minute! Dick Sargent…Dick York…Sergeant York….”
June 9th, 2008 at 10:05 pm
If I read this correctly, wouldnt “spooky action at a distance” allow us, in theory, to transfer information INSTANTLY? if to photons located say, a 100 lightyears apart does the “opposite coinflip” couldnt we then use that to signal 1′s and 0′s and thereby all the information in the world?
June 9th, 2008 at 10:11 pm
Uh Phil, ok — just as long as your play with the cat doesn’t involve a cyanide capsule. Wouldn’t want to have to call the ASPCA on ya.
June 9th, 2008 at 10:12 pm
Matthew Reynolds, I did say “Einstein called *this* spooky…” where the “this” means entanglement.
Gazza: yikes! I think I’ll wait until tomorrow’s coffee to tackle that.
Bruce: typo fixed, thanks.
June 9th, 2008 at 10:19 pm
Another interesting point here is that because you’re using an XOR key, there are multiple right answers for a given key. 00 or 11 will both give you a 1 and 01 or 10 will both give you a 0 (or vice versa, I get it mixed up. some comp sci major I was). This means that not only do you have a key that is unique, even if someone else did get the encrypted message and the key, they wouldn’t necessarily know that their decrypted message was correct. Fun with cryptography
.
June 9th, 2008 at 10:19 pm
You give the impression that the use of entanglement for encription is brand new. Seems like I read something about this at least a year ago, maybe in Scientific American. As you describe the use of it for criptic messaging, the piece I read also held great hope for unbreakable coded messages. If memory serves – and quite often doesn’t – the idea of some day using entanglement to “trasnport” matter across vast distances would be a way of insuring items were fathfully reassembled at the far end. We can only hope politicians don’t get a hold of entangled pairs, there is enough confusion in government now.
June 9th, 2008 at 10:20 pm
BicycleRepairMan: Unfortunately, no: when you measure the two photons your answers will be correlated, but you still don’t get to choose them (in the same way that I can’t communicate to you with purely random coin flips: I’d need some way to force the coin to land on heads or tails, which you can’t do here). So you can’t send information this way; you can just ensure that both sides have the same random information when they finally do measure. In many ways (but not all ways!) it’s as if both photons already “knew” whether they’d give a 0 or a 1, but were just waiting until you measured them. (When I say “not all ways,” though, I mean it: you can do experiments to show that the photons can’t possibly have this information stored in them. But the metaphor works, at least, in explaining why this kind of system can’t be used for faster-than-light communication!)
By the way, if you’re curious about quantum cryptography (and the history of cryptography in general), from what I remember Simon Singh’s “The Code Book” has a very good explanation — though GAZZA did a great job explaining the details that Phil left out of his post.
June 9th, 2008 at 10:22 pm
if a tree falls on my head while i am in the forest and no one is around to check on me, am i alive or dead
June 9th, 2008 at 10:25 pm
Hmmm, Pop, I can see why someone might reach that conclusion. In fact, my point was that this has been done over short distances, but never over long (1000 km) distances. From what I read in the linked article, that might distinguish between different interpretations (and I suspect it has commercial importance as well).
June 9th, 2008 at 10:27 pm
York is simultaneously Sargent/not Sargent?
If we had two Darrin Stevens a thousand 1k km apart….
Sneak out for a brew sometime Phil (et al)…walnut or sunshine?
June 9th, 2008 at 10:27 pm
Awesome post, Phil! This is exactly the kind of interesting stuff that keeps me a daily reader.
June 9th, 2008 at 10:28 pm
Ah, no Phil. Mr E had much better hair. Don’t kid yourself.
June 9th, 2008 at 10:29 pm
Also, the crypto geek in me can’t help but comment on one more thing: this system is often described as “unbreakable,” and as far as we know the “ideal” system is. But there was an interesting attack on this sort of system that was discovered, I think, by Adi Shamir.
As GAZZA explained, the security of the system relies on both parties setting up polarizing filters and keeping their orientation secret until after the photons are exchanged. But an attacker can wait until the filters have been set up (but before the photon has been sent through), and send *a lot* of light through. Some of it will bounce back off of the detector and back to the attacker — enough for the attacker to be able to measure the polarization of the filter. So while all of the theory that goes into this tells us that it’s unbreakable, the most obvious practical implementation of it *is* breakable. It’s for reasons like this that cryptography is really difficult to get right!
Anyway, I’ll stop hijacking this thread with crypto and let Phil get back to telling us about astronomy
June 9th, 2008 at 10:30 pm
So–in theory, if I log off and then log back on, there is at least a 50-50 chance of seeing Dick Sargent?
In the words of Mr R.B. Gumby:
MY BRAIN HURTS!
June 9th, 2008 at 10:50 pm
So you have three lasers making a pattern of bits randomly (or so we think). Laser one and two do not what each other are doing to each other but since there is a third laser in orbit which because of quantum entaglement is forced other into the opposite state of one of the lasers. After a while all the bits are transmitted to one of the stations and the two bits of infomation is compared. If laser A does not have an effect on the laser on the space station then it has to be B. Then the pattern is figured out (the key) and is transmitted to be B so in the future all you have to do is turn on the lasers on and control the pattern in on to send the message and therefore the second one will automatically pick it up and flicker a pattern and you can figure out what person at the first laser is trying to transmit. Am I even close to what they trying to do? And if this works you do not have to worry about somebodywith filters they won’t even know your sending the message.
June 9th, 2008 at 11:01 pm
I recall a really great article in Scientific American about this from January 2005, but it’s only available to subscribers. It included an illustrated inset with the same explanation Gazza gave.
http://www.sciam.com/sciammag/?contents=2005-01
However, I do not understand how quantum entanglement is involved.
I wonder, what happens if the “man in the middle attack” does not involve measuring the photons directly, but instead involves creating an entangled set of photons? Is this even possible?
June 9th, 2008 at 11:05 pm
Matt – interesting, I hadn’t heard of that. Off hand I would have thought that such an attack would be impossible to pull off undetected; on the other hand Adi Shamir is about a gazillion times smarter than I am, so I’m sure that he’s considered that.
I’ll have to hit Google.
June 9th, 2008 at 11:07 pm
“…then we have a code that is basically impregnable. No one else can tap into it, making it pretty secure.”
…
PRETTY secure, eh?
June 9th, 2008 at 11:11 pm
I thought I’d point out something before the woos get to this thread.
“Observe” isn’t meant as “human being looks at it and changes the world” sort of thing, but rather that an interaction took place that depended on the state of the system in question, which is what collapsed it. It wouldn’t matter if it was a human, a camera, or just some light that didn’t record the data at all.
If a tree falls in a forest and no one is around to hear it, it does make a sound.
June 9th, 2008 at 11:14 pm
This really sounds like the guy here in Squamish, that’s kind of a local celebrity, riding around on his bike, with a brim taped to his helmet that kind of makes it look like he is wearing the planet Saturn on his head. You know, Spooky at a distance.
June 9th, 2008 at 11:19 pm
Hi Phil,
If anyone claims to know what the quantum theory is all about, they haven’t understood it. Feynman.
L8R Winslow
June 10th, 2008 at 1:52 am
What GAZZA said is only part right.
The way the current quantum cryptography can be broken is a very simple one: The man-in-the-middle attack.
Say ISS is sending the key to A and B simultaneously. However, Mr. X manages to interrupt that transmission before either A or B receive it, and either sends it on after having read the key or sends his own keys to A and B, known to him, but not to the respective other party.
Now Mr. X can decrypt all the messages A and B send, and even send answers, without A or B being any the wiser.
This has been discussed on Bruce Schneier’s blog (www.schneier.com) at length, but sadly Mr. Zeilinger, who is the bigwig in quantum cryptography these days, seems to understand quantum (a lot better than me, I admit) but not basic information technology. I’ve been at a speech by him and that was what I took home from it…
June 10th, 2008 at 2:32 am
I work on this stuff, so I feel justified to say something about it (well, more justified than normal, anyway…).
Phil, your explanation of entanglement is somewhat incomplete as it stands. You consider only the “heads” and “tails” states of the coin, but in quantum mechanics you can also have the states “head+tail” and “head-tail” (that’s a minus, not a hyphen). These are the infamous superposition states. The crucial thing about quantum entanglement is that you not only have perfect correlations when you look at the heads/tails properties, but also when you look at the head+tail/head-tail properties.
For those not afraid of maths, this is the best (correct) way to look at it: heads (H) and tails (T) are two perpendicular vectors in a plane, both starting at the origin. The H+T and H-T states are also orthogonal and starting at the origin, but they are 45 degrees rotated with respect to the H and T vectors. You can decompose the H+T vector in terms of H and T components (or vice versa). A measurement is associated with a pair of orthogonal vectors, so you can choose to measure the H and T pair, or the H+T, H-T pair, but not both.
When you ask “what state is the coin in?” the answer is H+T, but when you measure the H,T pair you will find H or T each with 50:50 probability. This is what people mean when they say the coin is both heads and tails at the same time when the state is H+T or H-T.
Wow this is getting longer than I intended…
That is not quite true. Two-qubit entanglement (the example described here) is completely understood, and there are many more classes of entanglement that are well-understood, in the technical sense. It’s still weird from a classical point of view, but that’s not the same as “not understood”. The Feynman quote floated by one commenter is rather unfortunate in this respect.
Many people couple the photons into optical fibres in order to get around this. The quantum protocols work just as well, but due to attenuation in the fibre (photon loss) you are limited to a few hundred kilometres. The main advantage of using the ISS is that once you’re out of the atmosphere your photon losses are negligable.
June 10th, 2008 at 2:45 am
Quantum Entanglement is one of those subjects that you can get and totally not get at the same time. My brain loves this kind of workout even if it does leave me feeling like a load of grey goo is gonna leak outa my ear at the end of it. Thanks Phil
June 10th, 2008 at 3:54 am
There are commenters here that knows more about this than me, but I can’t resist pontificating on physics. Hopefully I will be corrected on any errors.
AFAIU, if you buy the decoherence picture, classical (i.e. large degree of freedom) systems emerge from QM. And decoherence is starting to get investigated seriously now. But, according to some iconoclasts such systems are possibly just effective approximations to QM systems, unresolved due to unavoidable real life coarse-graining during measurement.
Personally, I don’t think this question should be considered unresolved, as decoherence emerges as parsimonious.
But, who knows?
@ Matthew Reynolds:
I’m not sure which law you are referring to. Einstein was concerned with realism, and the measurable absence of local hidden variables contradicts local realism. This is a paradox only if you assume hidden variables, locality or realism. No laws are broken or contradicted.
Pieter Kok describes superposition states and measured states, as I believe this is what you are trying to get at.
@ John B. Sandlin:
The point is that there is no communication (or special relativity would be broken). There is only this correlation to know about as Phil describes, but it doesn’t help you to send actual information.
You don’t know any more or less about the system after the measurement, except that the superposition state has collapsed into one of the possible measurement states. It did it “here”, so it did it “there” (or rather “there” is “here” as far as the state was concerned).
The correlation is the same, and the outcome was, as per QM usual, stochastic. How do you make communication, send information, through a stochastic outcome?
You can instead ask about if a non-local entangled quantum system collapses instantaneously. Basic QM doesn’t predict a speed (delay time), and that is AFAIU consistent with observations so far. I don’t know if you could add such a description if necessary, I can’t immediately see why not, so perhaps it is wrong to say that QM depends on it.
June 10th, 2008 at 4:11 am
Einstein, smarter than me? Absolutely not. As a matter of fact, last night I falsified most of his theories, and I also came up with a pretty good proof that the Himalaya is actually several kilometers below sea level and, therefore, flooded. Too bad I was way to drunk to remember how these proofs actually went, but my audience was pretty impressed, so it must be true.
June 10th, 2008 at 4:19 am
jokergirl, I am pretty sure that the quantum cryptography cannot be defeated the way you describe. In fact as far as I know this is exactly the kind of attack that this system is intended to defeat.
If you interrupt the signal without knowing the orientation of the polarization filters you will cause a certain number of the photons to go “missing” when your filter is set opposite to A and B, and B will notice that it didn’t receive photons that it should have giving away your presense.
As for sending instantaneous communications using quantum entanglement, from what I understand the catch is that you need to know the space-time coordinates of both parties in order to makes sense of the collapsing waveforms which limits your communication to the speed of light.
June 10th, 2008 at 4:23 am
[...] I’m not recommending it yet because I’ve only checked it out for a couple of days, but today’s post is [...]
June 10th, 2008 at 4:53 am
Can it be that, in some other unknown dimension, entangled particles are actually in the same place? They may be miles apart in our perceived dimensions, but communicate instantaneously in another, where there is no distance between them at all.
I guess this is what the mathematicians are talking about in the 10 or 11 dimensions of M theory.
Talking of mathemaiticians, can anyone answer this?
“Once in my life, I dropped a nickel that came to rest on its side. So one-in-a-million may be accurate odds on that.”
Toss a long cylinder, like a pencil, and it will almost certainly com down on its side. Toss a short cylinder, like a coin, and it will alost certainly come down on one end. What are the dimensions of the cylinder which has equal probability of coming down heads, tails, and side.
June 10th, 2008 at 5:04 am
I see several other people have already mentioned Schrödinger’s cat, which I find freaky.
I had never even heard of Schrödinger until a couple of weeks ago. Then in the course of just a couple of days, both Big Bang Theory (a show that has thought me everything I know about physics) and Bones made references to it.
Forget Dick York. It’s me who’s in the frakkin’ Twilight Zone.
June 10th, 2008 at 5:08 am
Some system requirements…
* Windows XP or later or Mac OS X or later
* Access to IIS…
June 10th, 2008 at 5:08 am
That’s ISS.. Damn you, Microsoft..!
June 10th, 2008 at 6:02 am
Ok, having read the posts above it is obvious to me many people know more about this stuff than me. So here are some basic questions;
1. I see Quantum Mechanics, Quantum Theory, Quantum Entanglement, and Quantum Physics, and I am sure I missed a few other Quantums. I am assuming they are related, but are they just different pieces of the same pie, or does one term act as the study, or science, for all of them?
2. Half fun, half serious. In Star Trek, there are Quantum Torpedoes. Since science fiction like Star Trek and Star Wars tries to do a good job of basing future technology on present day theory, what is the potential for Quantum anything to be used as a source of energy? Or did the Star Trek writers just like the word Quantum and they stuck it in there?
June 10th, 2008 at 6:09 am
I should also point out that Phil explained it, from my point of view, great. I now have a sort of you know kinda somewhere vague barely holding on basic understanding of Quantum Mechanics. I also have a headache.
June 10th, 2008 at 6:13 am
If I may quote Tom Servo:
“AAAAAAAAAAAHHHHHHHHHHHHHHHHHHHHHH!”
*head erupts in a shower of sparks*
Thanks, Phil, you just blew my mind again. Now I have to go get a replacement. That’s the third one in a month!
June 10th, 2008 at 6:36 am
I too got into the trap of thinking that this allowed instantaneous information transfer, and I started warming up my teleporter, only to have someone ruin my little dream. Ah, well. Also, why isn’t teleporter in this dictionary?!
P.S. Einsteins with out a doubt had the best hairstyle anyone could hope for, and don’t you dare post anything to the contrary!
June 10th, 2008 at 6:42 am
I too both understand QE and I didn’t understand QE as well at the same time. If you were to ask me a question, then it will turn out that I didn’t understand it though, but that’s only a well known quantum effect
)
June 10th, 2008 at 6:50 am
QM is the clearest evidence we have that our universe is in fact a simulation running on the laptop of a moderately bright post-human physics graduate student. Quantum weirdness is actually the result of the self-maintenance of his laptop’s filesystem. We pause as it synchs.
June 10th, 2008 at 6:58 am
GAZZA did a great job explaining it.
I also would like to add that a quantum entanglement experiment was done last year between two neighbouring Canary islands, and it was determined that entanglement remains intact through 144 km of atmosphere. This means that this method can be used in communication with satellites in space, and that’s what they’ll be doing now using the ISS.
The article is here:
http://www.esa.int/esaCP/SEMXM7Q08ZE_index_0.html
June 10th, 2008 at 7:05 am
I figure the probability ought to be equal if the surface area of the side was equal to that of both ends. So that would be:
h2πr=2πr2
or
h=r
That’s a totally uneducated WAG, but it’d be neat if it was that clear and simple.
June 10th, 2008 at 7:13 am
“2. Half fun, half serious. In Star Trek, there are Quantum Torpedoes. Since science fiction like Star Trek and Star Wars tries to do a good job of basing future technology on present day theory, what is the potential for Quantum anything to be used as a source of energy? Or did the Star Trek writers just like the word Quantum and they stuck it in there?”
I remember seeing some behind the scenes stuff about how Star Trek got produced… and no. It has nothing to do with QM.
The writers (literally!) just pluck words from a technobabble book and plop them in the script wherever they fit.
June 10th, 2008 at 7:20 am
Figures. Thanks Tyler.
June 10th, 2008 at 7:23 am
No fair, you changed the outcome by measuring it!
June 10th, 2008 at 8:12 am
…huh? But if the coins were rigged, where’s the big bug here? It was a trick.
June 10th, 2008 at 8:16 am
Ah, I was actually thinking about htis last night whilst trying to get to sleep in bed! I was trying to relate things to QM… I came up with this, tell me if I’m on the right track or not:
You awaken in a white room with a door – the door is locked, can not be physically opened,smashed, etc.
There is also a ‘marker’ or ‘pen’ in the room with you, now;
You know that in this room, time is infinite – you dont need food/water/ etc (its a hyperthical situation)
Now, to open the door, a precise object or word or image must be drawn somehwere in the room – it needs to be PRECISE. you have NO knowledge of what this object is – but you know that you need to draw something in order for the door to open.
Now, lets say hte object that needs to be drawn is extremely complex, lets say ‘a precise schematic of the LHC’ or something equally as complicated.
The argument in QM favor would be that since I have an infinte amount of time, and no knowledge of the LHC – then I am certain to eventually draw the precise schematic and open the door.
Does that qualify as a QM analogy?
or was i just extremely tired
June 10th, 2008 at 8:35 am
I just read great book based on entanglement called (you guessed it) ‘entanglement’ by Michael Brooks who is the Physics consultant to ‘New Scientist’ magazine in London it’s fiction but a great read and a must for American readers I was late for work today cause I just couldn’t put it down !
June 10th, 2008 at 8:38 am
The main problem I try to get my head around is, how is one photon supposed to know it is “married” to another, particular photon?
June 10th, 2008 at 8:43 am
If it’s instantaneous over a vast distance that would also mean the information was traveling much faster than the speed of light.
June 10th, 2008 at 8:48 am
These kind of subjects remind me of Jim Carrey in the Truman Show, scratching around the edges of his reality. When you find stuff like this, that breaks with your normal perception of reality, you know you must be heading in the right direction.
June 10th, 2008 at 8:49 am
Ow. brain hurts now. thanks. BTW, Nice to see you in the new issue of geek magazine. My first issue arrived yesterday. The creepy part, Phil and Wil Wheaton… in the same issue…if I ripped a few pages out, Phil, the two of you could finally be together.
June 10th, 2008 at 8:57 am
@ Don McArthur:
We can only hope this simulation is not running any version of Windows! Especially VISTA!
June 10th, 2008 at 9:15 am
Can we take a quick break? It appears my intelligence circuits have melted after reading all that.
June 10th, 2008 at 9:26 am
So to apply all of this to a real world example that a layperson like me might understand, is the following correct:
I need to lay down now, before I take a quatum leap off of a cliff…
June 10th, 2008 at 9:28 am
…that would be a “quantum leap” not “quatum leap”. See what this has done to me? *sobs and looks for a cat to cuddle – I’m sure there was one around here earlier…*
June 10th, 2008 at 9:30 am
Different Matt here from the one that posted earlier. I have a question, though – since these entangled particles have a way of communicating their states faster than the speed of light, doesn’t that -imply- that there is some kind of physical mechanism that breaks the speed of light – and that this FTL communication could be used for, well communication? Of course I don’t know enough on modern QM at this point to know if that’s a valid question or not.
June 10th, 2008 at 9:52 am
riki, entanglement does not allow faster-than-light signalling. Suppose I have a photon that is entangled with someone else’s photon on the moon. A polarization measurement on my photon tells me whether it is horizontally (H) or vertically (V) polarized. The measurement outcome (H or V) is fundamentally unpredictable. So even though I know exactly how a measurement outcome (H or V) would change the state of the photon on the moon, I still don’t know which measurement outcome I will get, and neither does my buddy on the moon. I have to send a classical signal to the moon that tells him what my measurement outcome was.
This is not a proof, but it does offer some intuition as to why superluminal signalling is not possible. The formal proof is quite a bit harder.
Another thing to consider is that entanglement is also part of relativistic quantum field theory (this is the mathematical foundation of the standard model, and it covers everything quantum, short of string theory). Since this is a Lorentz covariant theory, faster-than-light signalling is implicitly prohibited.
June 10th, 2008 at 10:04 am
alfaniner, the marriage of protons is more or less a legal issue. First there is a constitutional amendment that a marriage shall be between a photon and its opposite. The particular photons are determined either by prearrangement at a very young age or by random meetings and choosing, as far as they know, on their own. Of course, it is not truly random as that might lead to same state photons becoming entangled. Of course, these are basic features of this wooniverse. Your dimension may vary.
June 10th, 2008 at 10:05 am
[...] Phil’s description of what’s going on, and comment 18. Maybe Brian can explain it more, I just think it’s [...]
June 10th, 2008 at 10:13 am
The only “particle” I know of that could possibly be the information carrier between two such quantum states is the tachyon, which posseses less energy(and infomation) the faster it travels. Thus, at a velocity just slightly less than infinity(sic) it could only carry the least amount of info possible, ie, either a particle is in one state or another.
See how easy quantum physics is?
Gary 7
June 10th, 2008 at 10:14 am
Everyone here (if they haven’t already) needs to pick up Neal Stephenson’s Cryptonomicon. My favorite book of all time. It deals a lot with cryto, math with a back drop of WWII, treasure hunters, espionage, etc. Incredibly funny and a highly recommended read.
June 10th, 2008 at 10:15 am
jokergirl, there is a difference beteen an Eavesdropper (Eve) and a Denial-of-Service (DoS) attack. Eve just tries to intercept the key whilst remaining undetected, whereas a DoS is set up to prevent communication between the two parties in the first place. Quantum cryptography protects against Eve, but not a DoS attack. I can assure you that Anton Zeilinger is well aware of this.
Another subtlety is the “authentication problem”. Suppose Alice and Bob try to generate a shared random key. How does Alice know she is talking to Bob and not to Eve? The quantum cryptography protocol discussed here does not protect from this type of attack either.
Some history: This quantum crypto protocol was invented by Bennett and Brassard in 1984, and elements of it date back to Wiesner, who was working on this topic in the early seventies. In 1991 an entanglement-based crypto protocol was invented by Ekert, and all this culminated in the invention of quantum teleportation in 1993. Teleportation now plays a central role in architectures for quantum computers.
June 10th, 2008 at 10:27 am
What I am considering though, people have stated that faster-than-light data transferral is still as impossible as I thought it was 5 minutes ago, however, despite US the observer not knowing the outcome, an outcome must still be reached, and thus something is transfered. When one particle spins up another spins down..instantly. They ‘talk’ to each other, faster than light, so even though we may not be able to utelise this, due to its unpredictability, the fact remains that info is still transferred..yes?
And I know the answer is inevitably going to be no, but I want to know why!
Also, forgive me if I am not making sense, but it is 2.30am here, I should be popping off to be right about now!
June 10th, 2008 at 10:38 am
Rowan, information transfer requires observers. When you’re not observing (that is, measuring) the photons it is meaningless to talk about information transfer.
June 10th, 2008 at 10:43 am
Faster than light?
Sounds like you can know immediately that the state changed, but not how it changed?
So could you set up a communication system that just measures state change times/rates and converts that into some sort of information stream?
Seems you could do this without caring about what states the entangled particles are in – ever. You just care if the states are changing because the information is riding on that.
June 10th, 2008 at 10:52 am
Mike R., exactly the point I was trying to make on the FTL communication thread on the forum.
So, even if it would break causality, it has not yet been disproven that FTL communication (or, to be really picky, instantaneous information transmittal) is possible with this method.
Not sure if we have the technology to “flip” yet, though. But the technology to “look” we have, and the current q-crypto method relies on detection of “looking”.
June 10th, 2008 at 10:53 am
It’s not FTL because the information does not “travel” by the way. At least not in our definition of timespace. It is instantaneous.
June 10th, 2008 at 11:14 am
From what I’ve seen here, that’s not possible either. Correct me if I’m wrong, but it seems to me that you can’t know that a state has “changed” without measuring it, and that’s entirely the point. Since measuring it CAUSES the outcome to collapse, there would be no way looking at it to know if it was your measurement that caused the collapse or if one was made to the particle a million lightyears away.
However, is there a way to cause a collapse in such a way that when it does collapse, that change causes some reaction which CAN be detected? In other words, the reverse?
Anyway, to the one above thinking they found a way to crack the quantum code, re-read the above description of how “man in the middle” is defeated. There is no way for the man in the middle to know the original orientation of the filters any more than the recipient, and since an extra file is sent telling some of the bits, even if they also intercepted that, it STILL doesn’t tell them which orientation of filters was originally used (see how the bits are originally derived for the reason why), so he still can’t perfectly replicate those first 50.
June 10th, 2008 at 11:21 am
Okay actually I’ve rethought this and tell me if I’m getting closer.
Entanglement only “preriggers” the photons to be measured one way or another when the time comes. Measuring one won’t even cause the other to collapse, it’ll just inform you what the other one has been predetermined to collapse into.
The problem is that I understand that the very aspect of the state being indetermined in a very real sense and not really predetermined at all is vital for physics to behave as we observe it, and if it actually was pre-set, we would get difference observations than what we do get. Is this then the one exception to the rule?
June 10th, 2008 at 11:28 am
But saying something is “instantaneous” automatically creates causality problems, doesn’t it? If this instantaneous exchange of information were to take place over several lights seconds, what’s to say you could then observe yourself receiving a signal before you actually sent it?
June 10th, 2008 at 11:35 am
I’ve been thinking about all of this, and now I have a headache. Does this mean that my twin brother also has a headache, or that he doesn’t? Or maybe he might do, if I call him and tell him I have a headache. Or if I don’t, he won’t. Am I getting there? Am I confused? And if I’m confused, does that mean that said twin bro isn’t, in which case he could explain all of this to me, and neither of us will have headaches? I’m going to lie down now, and play with my cat for a while. Then again, I haven’t seen the cat, so is he even alive? Or is he dead? Tell you what, forget the cat. I’m just going to lie down.
June 10th, 2008 at 11:40 am
“Does this mean that my twin brother also has a headache, or that he doesn’t? ”
In the alternative universe of mirror opposites, he doesn’t have a headache, for a change.
Or is he a she? The mind boggles.
I am fine with it, just so long as I get to have one of those mirror-thingys like “Bad” Kirk did when he visited the mirror universe.
June 10th, 2008 at 12:00 pm
Here is a classical analogy of “faster than light” correlations: Imagine a vase with two balls, one black, one white. Alice and Bob each blindly draw a ball and travel to opposite ends of the universe (so to speak).
Before Alice “measures the state of the ball”, that is, before she checks if it is white or black, she describes the colour of the ball as a probability distribution: white:black = 50:50. She also knows that the colour of her ball is correlated with the colour of Bob’s ball. When Alice looks at her ball, and discovers that it is white, she instantly knows that Bob’s ball is black. Did any information transfer take place? No! Is there any faster-than-light business going on? No! Bob is completely oblivious to Alice’s dealings with her ball.
The correlations in entanglement are very similar. As soon as you measure the polarization of an entangled photon you know instantly what the polarization of the other photon is, whatever pair of perpendicular polarization directions you choose. It is just that the quantum correlations are “stronger” than the classical correlations. It is an interesting fact of Nature that these quantum correlations allow for teleportation (which can’t be done classically), but still do not allow superluminal signalling.
For a proof that no superluminal signalling is possible in quantum mechanics, see Ghirardi, Rimini, and Weber, Lett. Nuovo Cimento Soc. Ital. Fis. 27, 293 (1980).
Irishscribe: no, your twin brother is just younger than you.
June 10th, 2008 at 1:05 pm
Pieter,
Am I correct in my thinking when I say that once the polarization state of an entangled photon is measured there can be no further manipulation of the photon? (Ie. you can’t then re-polarize it thus effecting the other photon). If that’s correct I believe that is the part many folks are missing in the FTL signalling discussion – you can’t change the state of your photon once it’s been measured therefore you cannot send information by twekaing the photons at each end. Their states, while unknown, are for all intents and purposes set when they are created and are simply awaiting observation to reveal that state.
Is that correct? Or did I miss something here (OK, I know I’ve missed volumes… but something obvious!)
June 10th, 2008 at 1:46 pm
Pieter, if the Alice and Bob scenario is correct then there is no mystery to wrap our heads around. It would just mean that the entangled items are two halves of a whole, so to speak. That hardly seems to rise to the level of brain scrambling that is going on here. That is, if I take the black ball and run without looking at it I certainly would not be amazed in any way because I open my hand and realize that the remaining ball is white. To put it another way, Einstein seems to have been at a level that he would not consider such a thing spooky so I suspect that this does not really explain the idea of entanglement.
June 10th, 2008 at 3:02 pm
Hm, how come every time quantum mechanics are mentioned, a connection to Albert Einstein is made, like “Einstein didn’t like it” etc.? Almost never is the guy even mentioned who actually came up with QM, Max Planck.
What’s this obsession with Einstein as if hew was the the only physicist ever? He developed the Theory of Relativity, as we all know, an earth-shattering discovery for sure, and he had a freaky hairstyle. Yet QM is just as earth-shattering, so please give the guy who came up with it some credit, too.
June 10th, 2008 at 3:22 pm
“since these entangled particles have a way of communicating their states faster than the speed of light, doesn’t that -imply- that there is some kind of physical mechanism that breaks the speed of light – and that this FTL communication could be used for, well communication?”
It’s not clear that they do. Consider the following analogy. You have a 10 cm yin-yang symbol made out of plastic. You cut the symbol along the boundary between yin and yang and throw each one into a random box without looking. You now have 2 boxes, one with the yin in it (black) and one with the yang in it (white) but you don’t know which box contains which.
You put one box on a rocket ship and send it to a moon base. At exactly 12:00 UTC you open both boxes. The “wave function” collapses. It turns out that the box on earth contained yin which means that the box on the moon contained yang. But this information was revealed in both locations with no delay time based on the speed of light (it should have taken 1.25 seconds for information from the earth to reach the moon). Is this an example of faster-than-light travel? I don’t think so. The information contained in the boxes was correlated at creation time and no new information needs to be sent from one to the other. Just like in the case of entangled photons, the two photons are quantumly correlated at creation time and they do not have to send any information between themselves later.
June 10th, 2008 at 3:30 pm
I love QM! These were the few classes that came most naturally to me in undergraduate and graduate levels. Then again, I had really good teachers for each.
Larissa
June 10th, 2008 at 3:35 pm
Arto7, The crucial difference between the balls and the entangled photons is that the balls can be measured only on the black/white property, while the polarization of the photons can be measured in the horizontal/vertical directions, and also in two perpendicular directions that are rotated 45 degrees from horizontal and vertical. Quantum entanglement requires correlations not just in the hoizontal/vertical directions (which, as you remark, is the same as the white and black balls), but also in the rotated directions. This is an extra requirement over the classical case, and that makes quantum entanglement different.
Stark, you’re pretty much correct. Once a photon is measured it is usually gone (the photon is absorbed by an electron, which is excited to the conduction band of a semiconductor and accelerated by a voltage, creating more conduction electrons and thus creating a measurable current), but even if you can keep the photon, the act of measurement disentangles it from its partner. Whatever you do to the photon after the measurement has absolutely no bearing on the photon far away.
George, Einstein is just the most famous critic of quantum mechanics, and he used entanglement to drive its weirdness to breaking point. Plus: he coined the term “spooky action at a distance”, loved by journalists and bloggers all over.
June 10th, 2008 at 4:00 pm
Hmmm… actually, quantum mechanics has MUCH to do with Einstein. It was him who postulated that light consists of individual quanta, when he explained the photoelectric effect.
It could be considered that the entire field of quantum physics was born from that.
June 10th, 2008 at 4:28 pm
Phil, having met you in person, I gotta go with the others – Old Al had *much* better hair.
June 10th, 2008 at 4:31 pm
I have a little problem with subjects like these. We’re talking about very complex stuff that has to be explained and explained again as the subject moves from mouth to mouth. What is left in the end is -probably- a very simplified image of something that is very different in reality.
I like Schroedinger’s cat. It is both dead and alive, until you look at it. At least, that’s the simplified explanation. Call me stupid, but I can think of a gazillion experiments where a cat is both dead and alive until somebody bothers to check the poor animal out. My gut feeling tells me that the cat was either dead or alive in the first place and that we couldn’t tell it by the method we used to predict its faith. Chances, probability and statistics are not hard facts.
Now let’s create some particles in a specialized process where we know the particles are related by certain properties because we want them to. But hey, don’t look at the actual values of these properties. As long as nobody has examined a property it could be, no, it HAS every value. Because we cannot determine values of properties by the methods we use.
I have a lottery ticket here. Tomorrow the winning combination will be announced. You can buy it for 100,000 Euro. As long as you don’t look at it or read the winning combination in the newspaper the ticket is also the winning one. Deal? You may want to sell it yourself for 200,000.
Back to earth/space. Yes, we could encrypt things using entanglement. It’s somehow like giving a friend a computer generated random password on a paper in an envelope and send him away with the instructions not to open it. As long as there is no communication this password cannot be cracked or guessed by anyone. Not for certain at least. But when it is used it is just a password c.q. the public key of a master algorithm.
If we could change spins on demand and observe the changes at entangled particles the story would change dramatically. I’ve read popular scientific articles claiming this has happened, but every time I ended up at the word ‘could’ as in ‘you could win the lottery’. Buy my ticket?
June 10th, 2008 at 6:45 pm
LaCreption:
I haven’t studied quantum mechanics yet, but here’s what I learnt from books I’ve read:
The Schroedinger’s Cat situation cannot exist in our world because the air molecules around the cat are going to “observe” it, causing to be either dead or alive, but not both. Also, the cat itself is an “observer”, so the “multi-state” situation does not apply here.
So is the lottery ticket example. Firstly, whether a “human” observes the winning combination doesn’t matter. If this winning combination were generated by some computerized pseudorandom generator, then it’s not quantum mechanically random at all. That number has already been recorded into the disk/RAM/newspaper and whether you look at it doesn’t matter now.
June 10th, 2008 at 7:23 pm
Delayed choice quantum eraser:
http://en.wikipedia.org/wiki/Delayed_choice_quantum_eraser
Brain go boom.
June 10th, 2008 at 11:37 pm
It’s all very well to say that if you were to send information over huge distances that it would have to travel faster than light,…
…but perhaps they aren’t travelling at all.
Perhaps on some level, they are not even apart!
June 11th, 2008 at 12:05 am
Ahh thanks Pieter and everyone else for clearing this up. The piece information that I was missing was that he photons are entangled only at creation, and cannot be manipulated after measurement. That’s definitely the key piece of info I’ve ben missing out on ’till now. QM is confusing only when it isn’t explained properly by a knowledgeable expert.
June 11th, 2008 at 3:36 am
Phil Plait = awesome.
June 11th, 2008 at 4:27 am
|||////++++++++0000011111111
June 11th, 2008 at 9:33 am
I’ve never fully understood the issue with spooky action at a distance. Why are we looking at it as a cause and effect relationship? Observing particle A to be, say ‘on’, causes particle B to be ‘off’? Isn’t their relationship set when they get entangled and we simply don’t know that relationship until we observe one of them? That would rule out faster-than-light communication. What am I missing?
June 11th, 2008 at 3:03 pm
[...] at Bad Astronomy, Phil Plait wrote about this on Monday. He offers a cool short-version explanation of the quantum mechanics involved, and his comment [...]
June 12th, 2008 at 8:20 am
The thought of quantum DRM (http://en.wikipedia.org/wiki/Digital_rights_management) disturbs me greatly. Fortunately, most media companies are wising up to the value of open formats. Hopefully by the time this is applied to media, they’ll say “Nah, we’re ok.”