Quantum test pricks uncertainty

I put this in coffee house as I am no doubt the least qualified person on this board to discuss the subject, but I find it interesting and I believe others will as well.

quote:

---

Pioneering experiments have cast doubt on a founding idea of the branch of physics called quantum mechanics.

​

The Heisenberg uncertainty principle is in part an embodiment of the idea that in the quantum world, the mere act of observing an event changes it.

​

But the idea had never been put to the test, and a team writing in Physical Review Letters says "weak measurements" prove the rule was never quite right.

​

That could play havoc with "uncrackable codes" of quantum cryptography.

​

Quantum mechanics has since its very inception raised a great many philosophical and metaphysical debates about the nature of nature itself.

​

Heisenberg's uncertainty principle, as it came to be known later, started as an assertion that when trying to measure one aspect of a particle precisely, say its position, experimenters would necessarily "blur out" the precision in its speed.

​

That raised the spectre of a physical world whose nature was, beyond some fundamental level, unknowable.

​

This problem with the act of measuring is not confined to the quantum world, explained senior author of the new study, Aephraim Steinberg of the University of Toronto.

​

"You find a similar thing with all sorts of waves," he told BBC News. "A more familiar example is sound: if you've listened to short clips of audio recordings you realise if they get too short you can't figure out what sound someone is making, say between a 'p' and a 'b'.

​

"If I really wanted to say as precisely as posible, 'when did you make that sound?', I wouldn't also be able to ask what sound it was, I'd need to listen to the whole recording."

​

The problem with Heisenberg's theory was that it vastly predated any experimental equipment or approaches that could test it at the quantum level: it had never been proven in the lab.

​

"Heisenberg had this intiuition about the way things ought to be, but he never really proved anything very strict about the value," said Prof Steinberg.

​

"Later on, people came up with the mathematical proof of the exact value."

​

'Full of uncertainty'
Prof Steinberg and his team are no stranger to bending quantum mechanics' rules; in 2011, they carried out a version of a classic experiment on photons - the smallest indivisible packets of light energy - that plotted out the ways in which they are both wave and particle, something the rules strictly preclude.

​

This time, they aimed to use so-called weak measurements on pairs of photons, putting into practice an idea first put forward in a 2010 paper in the New Journal of Physics.

​

Photons can be prepared in pairs which are inextricably tied to one another, in a delicate quantum state called entanglement, and the weak measurement idea is to infer information about them as they pass, before and after carrying out a formal measurement.

​

What the team found was that the act of measuring did not appreciably "blur out" what could be known about the pairs.

​

It remains true that there is a fundamental limit of knowability, but it appears that, in this case, just trying to look at nature does not add to that unavoidably hidden world.

​

Or, as the authors put it: "The quantum world is still full of uncertainty, but at least our attempts to look at it don't have to add as much uncertainty as we used to think!"

​

Whether the finding made much practical difference was an open question, said Prof Steinberg.

​

"The jury is still out on that. It's certainly more than a footnote in the textbooks; it will certainly change the way I teach quantum mechanics and I think a lot of textbooks.

​

"But there's actually a lot of technology that relies on quantum uncertainty now, and the main one is quantum cryptography - using quantum systems to convey our information securely - and that mostly boils down to the uncertainty principle."

---

Hi GDR, I don't count myself as qualified to discuss quantum mechanics either, but off the cuff it seems that this would seem to challenge entanglement rather than uncertainty per se.I also don't know if this is "old hat" in physics that is just being played.Enjoy

This is a cool bit of physics but it in no way changes Quantum Mechanics or the Heisenberg Uncertainty Principle. What it does show is that the rather nebulous explanations of the Uncertainty Principle as arising from physical restrictions on measurement, sadly used in many intro books on QM, are largely erroneous. And this is all for the good.

No, it doesn't actually have anything to say about entanglement. Entanglementand the HUP are both deep seated consequences of the quantum nature of reality. What they have challenged is the age old explanation of the HUP in terms of the limitations of measurement wthin classical physics.

As cavediver has said, there was an explanation of the uncertainty principle very common in introductions to QM for undergraduate students (although I don't think most textbooks since the 1990s have used it) and common in popular accounts. The explanation being that when you measure the position of a particle you "destroy" or decrease your knowledge about the momentum of the particle, since the equipment disturbs the particle.However it has been known for a long time that this is probably a poor way to think about/teach the consequences of the uncertainty principle. Especially since the late 1980s when we began to understand much more about quantum mechanics (the increase in understanding coming from measurement theory and quantum computing).The experiment above agrees with the uncertainty principle, but shows older explanations and understandings of its meaning and content to be incorrect.The uncertainty principle "really says" the following:You prepare millions of copies of a particle all in the same state (let's say they're all produced via some identical procedure to ensure they're in the same state).Then let's say you take half of the particles and measure their position (at some fixed point in time after they're released from where they're produced, let's say). Since quantum mechanics is probabilistic, you'll get a spread of values for the location of the particle, as every measurement will produce a different answer. This is unlike classical mechanics where several particles in the exact same state will always produce the same answer.If you compute the average position of the particles, then you can compute the average distance from the average. This number is the position uncertainty, labeled .Then with the other half of the collection of particles, you do the same thing, except with their momentum and work out the momentum uncertainty, labeled .The uncertainty principle says that no matter what state you choose to create the particles in, the product of these two uncertainties always obeys: So the uncertainties can only be reduced so far, since their product can never be less than .So this is a statement about the uncertainties in results from experiments due to the non-deterministic nature of particles in quantum mechanics.It's not about the experimental equipment interfering with or disturbing the particle, an explanation which is commonly seen.Entanglement, which is separate to this experiment, is on extremely solid ground, especially with the modern versions of Aspects experiments.Edited by Son Goku, : Square!

Thanks Son Goku Just for clarity, is the Δ_p number the squared momentum average distance?So these Δ numbers are like standard deviations squared (and you could compute the standard deviations from them if you know the number of particles) yes?Again, for clarity, what's h? I assume it's one of the Heisenberg numbers with units for (momentum x distance)^{^2}?Enjoy

The uncertainties are the standard deviations (i.e. the square root of the variances)I don't understand what you are trying to calculate using the number of particles. That number does not show up in the equation.h bar is Planks constant over 2*pi. The units are joule-sec.Edited by NoNukes, : Add units

Hi NoNukes That's not what the post said -- one was just squared and the other was or wasn't (ie not clear) Isn't standard deviation = {sum_(m=1→n)[ave - data_m]^{^2}/n}^{^1/2} ?(sorry, I haven't learned latex yet)Certainly using the standard deviation makes sense to me, I'm just looking for clarity that this is what is used (or something similar)EnjoyEdited by RAZD, : .

Hi RAZD,Thanks for reading, there's a mistake in the original post. It should read "the average distance from the average", no squaring. The two quantities are just the standard deviations of momentum and position, as you said. In fact the standard deviation is probably a better phrase to use, as uncertainty implies some limitation on the equipment, where as standard deviation emphasises (correctly) the purely statistical nature of the result. , said "h bar", is a constant with units Joules x Seconds, specifically: Edited by Son Goku, : No reason given.

The title of this post would probably be a better name for the result. The whole point being that even if you build equipment which doesn't disturb or influence the particles, you still get the product of the standard deviations being larger than a certain number.It's a statement about the statistics of measurements performed on a large collection (or ensemble, as is said in physics) of identical particles, all produced in the same way. It does not concern how measurement equipment influences or disturbs a single particle when it measures it.Heisenberg had the original idea of equipment disturbing the particle, however it was Earle Kennard (in 1927) who derived the relation in its correct form first and interpreted it correctly.There are similar results for other physical quantities. For example if are the angular momentum around each axis (x, y and z), then the standard deviations obey: Where, is the standard deviation for angular momentum around the x-axis. is the standard deviation for angular momentum around the y-axis. is the average value of angular momentum around the z-axis and just indicates that you ignore the sign of this value, i.e. If the value is -0.4, you just use 0.4

At the risk of dating myself, I seem to recall the "observation principle" explanation being brought up in my undergraduate class. I cannot recall if the explanation actually appeared in the textbooks, and by the time I finished undergrad I had rid myself (mostly) of that erroneous notion. This reminder from Son Goku and cave diver may have finished the job.In defense of those older text books, Heisenberg used the observation principle explanation, and as I recall, forms of the observation principle formed a part some of those thought experiments that Einstein posed to Bohr. So anyone who makes the mistake is at least in good company.

Yes, but n is only helpful as part of the formula for calculating the standard deviations. If you are doing measurements it is only necessary that n be sufficiently large to make the calculation of std deviation meaningful.The experts have already addressed the rest of your question.