Question about Virtual Particles

[haushofer]

Perhaps people, including the popular literature, should stress more that when talking about QM or QFT, metaphores/analogies are used to translate the math into tangeble concepts.

 

[PeterDonis]

"QFT for the gifted amateur" by Lancaster, sidenote 2 in page 348

I don't have this textbook handy but I'll see if I can get hold of a copy.

You can think heuristically of an 'on shell particle slightly off shell' as a renormalized particle i.e.: a "dressed particle" in the sense that it is obtained from a "bare particle" (on shell) with the radiative corrections by virtual particles(off shell).

Except that a bare particle does not have to be on shell, since bare particles can correspond to internal lines, and a dressed particle that corresponds to an external line is on shell once you realize that the physically measured mass is the renormalized mass, not the bare mass.

 

[TrickyDicky]

Except that a bare particle does not have to be on shell, since bare particles can correspond to internal lines, and a dressed particle that corresponds to an external line is on shell once you realize that the physically measured mass is the renormalized mass, not the bare mass.

I wasn't implying that a bare particle has to be on shell, as a matter of fact that is why I referred to the renormalized dressed particle, because after renormalization one can no longer refer to on and of shell in the same way we do with the pre-renormalized Feynman diagram. See the definition of dressed particle in wikipedia:"dressed particle refers to a bare particle together with some excitations of other quantum fields that are physically inseparable from the bare particle. For example, a dressed electron include the chaotic dynamics of electron-positron pairs and photons surrounding the original electron." I figured it might aid as a graphical heuristic to get an idea of what "slightly off shell" means, if it doesn't work for you or find it misleading just disregard it.

 

[JK423]

TrickyDicky and PeterDonis are you still disagreeing on the reality of virtual particles or on the definions of off/on shell?

 

[yoron]

No - because virtual particles don't actually exist. They are simply an artefact of the mathematical methods used called perturbation theory and something called a Dyson Series:
http://en.wikipedia.org/wiki/Dyson_series

Thanks
Bill

I'm with you bhobba, virtual particles are a artifact. you're not the only one reaching that conclusion. And it makes a lot of sense to me, it simplifies.

 

[yoron]

Yes but there are no apparent forces in GR, at least that's the only way I can interpret the "appear". A thermal bath can't just appear, either it's there or it isn't. Otherwise you may be appearing to me this moment and fall into a Jabberwock if I accelerated in a certain way wrt you.

Relativity is based on observer dependencies, it's not based on gold standards.

 

[Gerinski]

So quantum fluctuations are real and they have observable properties similar to as if they were particles, it's just that they should not be called particles in the popular literature

 

[TrickyDicky]

TrickyDicky and PeterDonis are you still disagreeing on the reality of virtual particles or on the definions of off/on shell?

Never disagreed on the reality of virtual particles that I'm aware of; wrt on/off shell I think it is just an issue about words. QFT is admittedly hard to translate to english.

 

[ddd123]

Can't we reverse the point and consider "particles" are just another way of quantum modes to manifest? Otherwise we should start saying: photons are one thing, electron-proton binding in a hydrogen atom completely another thing - not so good. Entia non sunt multiplicanda praeter necessitatem.

 

[PeterDonis]

TrickyDicky and PeterDonis are you still disagreeing on the reality of virtual particles or on the definions of off/on shell?

I think it's more a question of how to translate the math into ordinary language, as TrickyDicky said. The ultimate answer to that is that scientific theories are not formulated in ordinary language; they are formulated in math. The math involved is unambiguous (though there are differences of opinion on how rigorous it is). But not all questions we can ask in ordinary language necessarily have answers in the math. For example, the question "are virtual particles real?" doesn't have an answer in the math; there's no mathematical property that corresponds to "real" or "not real". My personal preference is to stop asking those kinds of questions, but not everybody wants to do that.

 

[Vanadium 50]

For example, the question "are virtual particles real?" doesn't have an answer in the math

No, but questions like "can they be counted?" have answers, and I think most people would argue that counting is a property one would like real things to possess.

 

[bhobba]

So quantum fluctuations are real and they have observable properties similar to as if they were particles, it's just that they should not be called particles in the popular literature

How you reach that conclusion from this thread beats me.

If they were real all ways of doing the QFT calculations would have them - but they don't.

Thanks
Bill

 

[bhobba]

My personal preference is to stop asking those kinds of questions, but not everybody wants to do that.

I think only some want to ask that. As you understand the mathematical language of the more advanced areas of physics better you recognise the math is the theory and in math you can't express questions like that. Also very deep ideas like Noethers Theorem and the role of symmetry simply can't be expressed properly otherwise.

Thanks
Bill

 

[yoron]

Actually I do think you have a excellent point in lifting up 'thermal baths' ddd, " A thermal bath can't just appear, either it's there or it isn't. Otherwise you may be appearing to me this moment and fall into a Jabberwock if I accelerated in a certain way wrt you."

and one I'm wondering about too. But to me it is about causality? Observer dependencies suddenly seems only half of the answer.

"
Causality means that cause precedes effect : an ordering in time
which every observer agrees upon.

Q: Whether it is possible to change the order of cause and effect just
by viewing two events from a different frame.

A: two events can only be cause and effect if they can be connected to
one another by something moving at speed less than or equal to the
speed of light.

Two such events are said to be causally connected."

In a wider context causality is, to me at least, what joins the universe we see, no matter (local) observer dependencies? Seems to belong in the same category as tachyons to me, when thinking of it from causality? Maybe it's not belonging to this thread though.

 

[JVM]

Since QFT is simply a perturbation theory it means that we don't know what happens in these intermediate stages of a process, and although we use these intermediate processes, it is definitely what actually happens. The problem is that we don't have an exact description, QFT at this moment is just an approximation.

Maybe you can compare it with a Taylor series, if you're not really home in the subject, where a function may be described by some finite sum up to a certain approximation, it does not mean that the function is equal to this finite sum.
In QFT a process is approximated by some individual feynman diagrams, but this is only an approximation and is definitely not the real process.

Hopefully someone will find out what really happens, until then we are sadly limited to these approximations (which are incredibly correct in describing our world, and are not to be underestimated).

 

[PeterDonis]

A thermal bath can't just appear, either it's there or it isn't.

As the Unruh effect shows, this is false, however plausible it seems. The key point is that the notion of "a particle being present" turns out to be observer-dependent when quantum effects are taken into account; a quantum field state that looks like a vacuum to one observer (a freely falling one) can look like a thermal state with particles in it to another observer (an accelerated one). But it's the same underlying quantum field state in both cases. So your intuition here is simply wrong, however plausible it seems.

Whether it is possible to change the order of cause and effect just by viewing two events from a different frame.

In QFT, that's not quite how to put it. The way to put it is this: quantum field operators must commute at spacelike separated events. In other words, the results of quantum measurements at spacelike separated events cannot depend on which order the measurements are made in. The results can only depend on the order of measurements if the events are timelike or null separated--i.e., if the order of events is invariant.

 

[Jilang]

@ JVM Are there not many ways to skin a cat? Would it not be conceivable that each mode did contribute in the Many Worlds, so each mode is a real occurrence in a specific world?

 

[yoron]

Thanks Peter. Although the first citation wasn't mine it caught something I wondered about when it came to causality. The other citation I used came from Causality in special relativity

 

[Vanadium 50]

The key point is that the notion of "a particle being present" turns out to be observer-dependent when quantum effects are taken into account

Yes, but the notion of "a particle is registered in a detector" turns out not to be observer-dependent. Much of this confusion comes from asking questions that are not directly connected to experiment.

 

[PeterDonis]

the notion of "a particle is registered in a detector" turns out not to be observer-dependent.

More precisely, the notion of "a state transition took place" turns out not to be observer-dependent. But observers can still disagree on how to describe the state transition. The accelerated observer describes it as the absorption of a particle by the detector, changing the detector's state, and interprets this as the detection of a particle. But the inertial observer describes it as the emission of a particle by the detector, changing the detector's state.

 

[king vitamin]

In addition to what's been said about the anomalous/off-shell dispersions of virtual particles, I'd like to add that the virtual particle content is gauge dependent. That is, depending on the your gauge choice you can have extra virtual particles (ghosts or spurious Golstones) which are not seen in any physical setting. You're free to choose whether these virtual particles appear in your perturbative expansion or not. Certainly at least these are uncontroversially considered unphysical?

 

[haushofer]

Actually I do think you have a excellent point in lifting up 'thermal baths' ddd, " A thermal bath can't just appear, either it's there or it isn't.

A thermal bath consists of particles. These particles can be counted by a number operator as for the harmonic oscillator. However, this operator is only invariant with respect to inertial observers. Hence, accelerating observers will count a different amount of particles. See e.g. Nikolic's "myths and facts"-paper, which has an excellent treatment on this.

 

[ddd123]

Imagine a box that counts particles and displays how many it detects in a bright LED. Observers, both accelerated and non-accelerated will agree on the number displayed on the LED. They may well disagree on the source or histories of the particles, but there is no dispute as to the number.

A thermal bath consists of particles. These particles can be counted by a number operator as for the harmonic oscillator. However, this operator is only invariant with respect to inertial observers. Hence, accelerating observers will count a different amount of particles. See e.g. Nikolic's "myths and facts"-paper, which has an excellent treatment on this.

Which of these is correct? Maybe rather than strict number of particles, the invariant is the total energy that comes to the detector? Otherwise i'd be tempted to start taking the vacuum energy more seriously.

 

[Vanadium 50]

The number of particles that the detector registers is invariant. Different observers will disagree on the source and history of these particles, but the number is constant. This is exactly analogous to the relativity of simultanity, where different observers disagree about the time ordering of distant events.

 

[ddd123]

I take it to mean that also the kind of particles is invariant?

 

[PeterDonis]

The number of particles that the detector registers is invariant.

More precisely, the number of state transitions is invariant. But, as I noted before, different observers will have different descriptions of what these state transitions consist of (detection of particles vs. emission of particles).

 

[carllooper]

What can be observed of the physical world (and said about such) is far more important than the physical world in itself. Basically because there is no such thing as the physical world in itself. The physical world is always related to what is observed (and said about such). And vice versa. Otherwise how would we know that what was observed (and said) was even related to the physical world, let alone being of the physical world.

Now the moment you sever this connection between observations and the physical world, one or the other disappears. One is left with either an observation (in search of an explanation) or an explanation (in search of an observation).

Many theoretical physicists prefer the latter situation - maintaining the notion of a physical world (it's explanation) without the need to involve observations. For such theorists, observations are just questions (not answers). However, without an observation, the physical world becomes no different from the mathematical concepts that would otherwise "explain" an observation. There's nothing wrong with this idea but it's questionable whether it's really physics. However we can maintain it's at least a rehersal.

For example, one could program a computer, using the developed explanations (the mathematical models), to display an animation of some physical process. But the result is not itself the physical process. Its an animation of the physical process. A representation. A rehersal. Until the explanation is used to manipulate the actual physical world in some way, it remains somewhat removed from the physical world - living in it's own little bubble of sorts. Its not a bad bubble and there's all sorts of problems to be solved in that context.

But it's really important to connect an explanation back to the observations it otherwise "explains" - which means doing physical experiments - not just animations. Seeing if the physical world, when reconnected to observations, behaves in the way that is otherwise modeled or represented (eg. by an animation).

Of course, most of these experiments have already been done - but that's no reason not to do them again - precisely because, when they're not actually done, the explanation ends up back in it's own little bubble again, going nowhere, doing nothing but encouraging the perverse idea that the physical world really is no more than the explanation. For example, the idea that the physical world is fundamentally mathematical, has it's origin in this sort of bubble.

But an explanation only makes sense when there is something to be explained. Of course that's no reason to stop rehersing explanations. We never know when it will be needed. We want to be sure we have our game hat on when a real world problem is to be solved. Rather, it is say that an explanation can be not the entire truth - that there is a world bigger than the explanation - and that this always involves observations (actual experimental observations - not just the idea of such). That the physical world is not just the "physical world" of which the theorist prefers (that which can be modeled on a computer), but that real physical world which actually takes place and is expressed in actual physical experiments.

Anyway all of this is to suggest why explanations, on their own, can't be nailed down in terms of which is correct and which isn't. They are somewhat impervious to criticism - mathematical models particularly so. They will be correct regardless of the real observable physical world.

C

 

[Gerinski]

How you reach that conclusion from this thread beats me.

If they were real all ways of doing the QFT calculations would have them - but they don't.

Thanks
Bill

Well, there are real phenomena which get commonly attributed to virtual particle interactions (Casimir effect, vacuum polarization, the contribution to mass of the vacuum fluctuations etc). If they are real phenomena they must be caused by something real. Mathematical tricks can not produce observable effects can they?
If they all have an uncontroversial explanation in terms of real particles and fields only, kindly provide these. I don't know why would anyone bother with explaining them in terms of virtual particles if there was a more straightforward explanation available. That's why I said: "there must be something real going on causing these effects, perhaps the issue is that whatever it is, it should not be called 'particles', since as the thread seems to confirm, it has little to do with real, on-shell particles".

 

[Gerinski]

Advise:

first: study it all in its mathematical rigorous (if possible) formulation (it will take many many long years of hard study).

second: only after that, read popularizations (pop-science books, famous quotes, whatever...) if you like, but now knowing perfectly well what actually lies beneath.If you ever want to (really) understand some of it, never start with pop-books.

So you are asking all of us laymen to stay away from PF?
That's why some of us come here, to ask and to learn from people who know the business, to get them correcting our misconceptions and to grow in our understanding of physics. If you don't like sharing your knowledge with laymen that's fine, but maybe there are others who don't mind.

 

[bhobba]

Well, there are real phenomena which get commonly attributed to virtual particle interactions (Casimir effect, vacuum polarization, the contribution to mass of the vacuum fluctuations etc).

Haven't you been reading the thread? That they are caused by virtual particles is simply due to the pertubative formalism used. There are other ways of doing the calculations where virtual particles do not appear. Hence the cause can't be virtual particles.

Those other methods are mathematically more difficult requiring some really really advanced math and/or are only solvable on computers. That's why they are not used and pertubative theory used instead.

Here is an example of a reference:
https://books.google.com.au/books?id=bHtpAgAAQBAJ

It was also mentioned that Wienbergs 'bible' on QFT, even though it also uses perturbation theory, doesn't mention virtual particles either:
https://www.amazon.com/dp/0521670535/?tag=pfamazon01-20

Why isn't it done that way? It hard, very very hard. Those texts are known as 'challenging' to say the least, meaning basically for the very advanced - and even then only 'brave' ones.

All this thread is doing is going over the same thing again and again. Doing that will not change anything.

Added Later:
I dug up the following that hopefully will settle the issue:
https://www.physicsforums.com/threads/non-perturbative-qft-without-virtual-particles.485597/

Thanks
Bill

 

[ddd123]

Bhobba, since the unsettling part of all these posters seems to be the lack of real-ness of mathematical formalisms vs. the real effect they are associated with, why don't you answer them that the real cause of the real effect is simply the quantum field (modes)? Is it simply a coincidence or is there a reason why you don't want to answer in this way?

 

[Gerinski]

Bhobba, since the unsettling part of all these posters seems to be the lack of real-ness of mathematical formalisms vs. the real effect they are associated with, why don't you answer them that the real cause of the real effect is simply the quantum field (modes)? Is it simply a coincidence or is there a reason why you don't want to answer in this way?

Indeed that's just what I was asking. If there are real effects and you say that their cause are not virtual particles since they are nothing real, just kindly explain me what is the real cause of those effects and I will be happy. Again that's why I said: "there's something real some people call virtual particles even if they are not virtual particles. Just explain me what it is". If so the problem is just in the name given to the cause, the cause is still something real, isn't it?

 

[Jilang]

Oscillations of quantum fields that just don't have enough energy to make a real standalone particle that can be detected?

 

[craigi]

Oscillations of quantum fields that just don't have enough energy to make a real standalone particle that can be detected?

The Vacuum Energy is actually large. If we estimate it from the cosmological constant we get 10^-9 Joules/m³. If we calculate it from QFT we get 10¹¹³ Joules/m³. A very large difference, but in either case, there's plenty of energy for real particles, under the right conditions. For comparion, the energy for photons in the visible part of the spectrum is of the order of 10⁻¹⁹ Joules. We can go many orders of magnitude lower as we move towards radio frequencies.

Read the paper which I posted towards the start of the thread, to better understand the conditions in which we can generate real particles from the vacuum.

 

[Gerinski]

Haven't you been reading the thread? That they are caused by virtual particles is simply due to the pertubative formalism used. There are other ways of doing the calculations where virtual particles do not appear. Hence the cause can't be virtual particles.

I do really believe you, don't get me wrong, I am a layman nobody and you clearly know the stuff. I just want to say that this argument is weak. The fact that there are different calculation approaches or interpretations for something does not 'per se' imply that particularly one of them is the wrong one. Most probably they are all wrong in one way or another. The double slit experiment can be explained by Many Worlds, and there are other ways to explain it which do not invoke other worlds. This does not automatically imply that the Many Worlds are not real and the other explanations are the correct ones. Thus the sentence "there are other ways of doing the calculations where virtual particles do not appear hence the cause can't be virtual particles" does not seem a very definitive argument. Still, I do believe you, I just wish you could explain me what is it that causes the effects attributed in the perturbative formalism to virtual particles. Cheers.