Do virtual particle pairs interact gravitationally?

[bahamagreen]

When two virtual particles appear through pair creation, does their brief presence present a gravitational influence on the surrounding universe?
If so, then after they self annihilate, does that gravitational influence remain or disappear?

The external gravitational influence of black holes is often attributed to the mass that was present before collapsing to the singularity... does the gravitational influence of virtual particles remain in a similar way after they are "gone"?

And if so, how much "extra" gravitational influence in the universe might be attributed to this vs dark matter? It is usually posed that observations imply missing mass for the gravitation observed, but might it just as well be "too much" gravity not associated with existent matter?

 

[Simon Bridge]

When two virtual particles appear through pair creation, does their brief presence present a gravitational influence on the surrounding universe?

yes ... note: particles from pair production are not virtual.

If so, then after they self annihilate, does that gravitational influence remain or disappear?

It stays ... you need general relativity. Mass is energy and the energy is conserved ... the annihilated particles' energy continues as a photon. (Gravity of photons requires quantum gravity - which we don't have but are working on ... see link.)

The external gravitational influence of black holes is often attributed to the mass that was present before collapsing to the singularity... does the gravitational influence of virtual particles remain in a similar way after they are "gone"?

No. Neither the pre-black-hole star nor the particles are "gone" in this sense.

https://www.physicsforums.com/showthread.php?t=442266

 

[Chopin]

yes ... note: particles from pair production are not virtual.

I think he was referring to vacuum fluctuations, in which case the particles would indeed be virtual, right?

 

[bahamagreen]

Yes, vacuum fluctuations... and since this is happening everywhere all the time, if the gravitational influence remains, it seems there would be a kind of accumulation of "extra gravity" not accountable to present mass...?

Am I overestimating the magnitude of this effect?
It would seem by HUP to have to be universal in space and time... at the resolution of the Planck length literally "everywhere" and continuously "always" for the last 13B years or so...?

Even the lowest possible vacuum energy level must have these fluctuations, yes?

 

[tom.stoer]

When two virtual particles appear ..., does their brief presence present a gravitational influence on the surrounding universe?

One might think that the cosmologicakl constant is due to repulsive gravity due to these vacuum fluctuations. But up to know attempts to calculate these effects have failed.

One might try to quantize gravity perturbatively; then there would be virtual gravitons producing gravity, and virtual particles could exchange virtual gravitons as well. Unfortunately this approach fails due to non-renormalizibility of perturbative gravity.

 

[Vanadium 50]

This whole thing starts on a faulty premise. There is no "when". You can't say "there used to not be a virtual particle pair here and now there is". That ascribes properties to virtual particles that they do not have: it's like asking what is the color of the invisible man's hair.

 

[Bill_K]

o The source of gravity is the stress-energy tensor, and it is locally conserved. You simply cannot create energy and/or destroy it, even in a virtual process, and despite whatever you think you know about the Heisenberg uncertainty principle. There is no cosmic bank account where energy can be borrowed, even short-term. So virtual particle pairs cause no gravitational disturbance.

o The vacuum state is time-independent. Remarks you see in the popular press about virtual particles "popping in and out", or having a "fleeting existence" is an attempt to picture classically what is essentially a quantum feature - the vacuum is a superposition of states with various virtual particles present. But they are always there - they don't pop in and out!

o Virtual particles may be off the mass shell. This means that an electron-positron pair may be present without costing the 1.02 MeV that a real pair would have. The energy of a virtual particle may be positive, or zero, or even negative. Since the energy of the vacuum is zero, the virtual particles must add up to zero energy - if one virtual particle has positive energy, some other virtual particle will have negative energy to make up for it.

 

[bahamagreen]

I seem to be reading both yes and no answers to these questions:

Is their existence temporary?
Simon Bridge - yes
Vanadium 50 - no
Bill_K - no
Wiki - yes

Do they have a gravitational influence?
Simon Bridge - yes
tom.stoer - no
Bill_K - no
Wiki - looks like yes, but not listed in the field interactions

If so, does this influence remain?
Simon Bridge - yes
Wiki - can't tell

Wikipedia's Virtual Particle page's first paragraph states,

"In physics, a virtual particle is a particle that exists for a limited time and space. The energy and momentum of a virtual particle are uncertain according to the uncertainty principle. The degree of uncertainty of each is inversely proportional to time duration (for energy) or to position span (for momentum)."

In the Manifestations section it states that,

"For the gravitational and electromagnetic forces, the zero rest-mass of the associated boson particle permits long-range forces to be mediated by virtual particles."

But then lists a dozen "field interactions which may be seen in terms of virtual particles" none of which seem to produce a gravitational influence.

 

[Bill_K]

Is their existence temporary? Bill_K - no

Their range is limited by the fact that they are off the mass shell. Temporary? No.

Do they have a gravitational influence? Bill_K - no

Everything that carries energy interacts with gravity, so in that sense they "have a gravitational influence." Their presence does not cause the interaction to change.

Wikipedia's Virtual Particle page's first paragraph states,
"In physics, a virtual particle is a particle that exists for a limited time and space. The energy and momentum of a virtual particle are uncertain according to the uncertainty principle. The degree of uncertainty of each is inversely proportional to time duration (for energy) or to position span (for momentum)."

Often the energy and momentum of a virtual particle is summed over, so in that sense only it is undetermined. Wikipedia's mention of the uncertainty principle in this context is unfortunate, and mistaken.

 

[Simon Bridge]

I seem to be reading both yes and no answers to these questions:

Read more carefully ... we are not disagreeing - we are talking about different interpretations of your question. For instance: I was talking about pair production or real particles. Turns out you weren't talking about that.

I don't think the usual vacuum fluctuation model has a gravitational component (iirc: we need quantum gravity to pull this off) ... can someone clarify this for me?

I would maintain that virtual particles are an artifact of perturbation theory[*] ... steps in a calculation. If the underlying equation being solved includes gravitational effects then the intermediate steps would be expected to reflect this. If not then no.

... either way, however, current ideas do not lead us to expect that we'd find any such cosmological contribution from them.

-------------------------

[*] the wikipedia article takes the other POV: that conservation of energy can be violated for sufficiently short times... allowing quite massive particles to shift energy and momentum around short distances.

 

[tom.stoer]

I don't think the usual vacuum fluctuation model has a gravitational component

Drawing vacuum bubbles with virtual graviton exchange you see the 'gravitational component'. But it's neither clear whether this results in something like 'gravity', nor can it be made consistent mathematically (afaik). Therefore this question cannot be answered unless we have a consistent theory of quantum gravity.

I would maintain that virtual particles are an artifact of perturbation theory[*] ...
-------------------------

[*] the wikipedia article takes the other POV: that conservation of energy can be violated for sufficiently short times... allowing quite massive particles to shift energy and momentum around short distances.

Simon is right and wiki is wrong! Energy and momentum are conserved exactly. In perturbation theory energy and momentum are conserved at each vertex. In addition the two equations [H,P] = 0 and [H,H] = 0 should be valid, therefore energy and momentum are conserved. But one should keep in min that this is a statement regarding QFT w/o QG.

 

[Bill_K]

I would maintain that virtual particles are an artifact of perturbation theory

This viewpoint would make more sense if it were not for the fact that virtual particles and real particles form a continuum. Where do you draw the line and say, this one is an artifact, but that one is real?

Any particle with a finite lifetime also has a finite width, and consequently ventures off the mass shell. Do you consider just those particles that live long enough to form a visible track real, and the others artifacts? Is a W-boson, with a lifetime of 10^-25 sec and a width of 2 GeV merely "an artifact of perturbation theory"? Or a photon emitted from the sun and absorbed 8 minutes later on Earth - slightly off the mass shell. Real or artifact?

 

[tom.stoer]

technically speaking real particles are Hilbert space states whereas virtual particles are propagators

 

[Bill_K]

Ok, but even if you don't do perturbation theory, you still have propagators, don't you? The place where perturbation theory comes in is when you expand a full propagator in terms of free propagators. So it's not virtual particles that are the mathematical artifacts, it's the bare virtual particles.

 

[tom.stoer]

But a propagator and a Hilbert space state are always different mathematical entities

 

[Simon Bridge]

I would have taken as a rule of thumb that the real particle is the one that we detect in Nature. The rest are some sort of math.

Has anyone detected the particle-pairs from vacuum fluctuations?

It is an artifact of the calculation process when it is something we calculated ... ergo, it has whatever properties that we allow it to have within the model we are using to do the calculation in.

We do this in classical physics when we, say, neglect the gravitational force between two charges while still allowing them to have inertia. Ergo: Coulomb charges don't have gravity... but real-life electrons and protons do.

The virtual particles of Field Theory are a bit more complicated than that though aren't they? They occur as intermediate steps in a calculation which start and end with real particles. However, we don't just get to use any old virtual particles in the intermediate steps, and we can produce them in particle accelerators and, again, not just any old mass particles. (How it is that you cannot pair-produce matter with any arbitrary mass is for another thread.)

Perhaps it would help to examine what may happen in terms of gravity in something like the beta decay of a neutron in free space? Looking at the intermediate stage you have a large mass gain in the W boson, lasting for a short time ... a student may imagine that this would result in a small gravity pulse.

Presumably the real, in-Nature, particle, with mass, has a gravitational effect. The Universe can work this out because the Nature knows more physics than we do right? Maybe Nature has a working version of quantum gravity or maybe it's something else.

The question, though, was about the cosmological significance of vacuum fluctuation particles having gravity.

 

[Bill_K]

I would have taken as a rule of thumb that the real particle is the one that we detect in Nature. The rest are some sort of math.

Ok, that definition excludes many things! Neutral pions are not real, since their lifetime of 10^-17 sec is too short, and we can see only their decay products. W and Z mesons, and Higgs are not real, for the same reason. Won't the guys at CERN be disappointed. And any neutral particle whose presence is only known through missing mass. Quarks and gluons too go in the bin, having been deduced indirectly but never detected. None of these are real particles, by your definition, only math.

Has anyone detected the particle-pairs from vacuum fluctuations?

Isn't that what the Casimir Effect is about?

The virtual particles of Field Theory ... occur as intermediate steps in a calculation which start and end with real particles. However, we don't just get to use any old virtual particles in the intermediate steps

This is getting close to the point I am making. The virtual particles are the SAME as "real" particles, just with a different mass/momentum.

(EDIT: With the exception of ghosts, which I grant are fictitious.)

Obviously you can't reach inside a Feynman diagram and detect one of the intermediate particles (whatever "detect" means). But you can do something very similar. You can select a subgraph of the diagram, and by merely adjusting the momenta of its legs, make it "real".

Perhaps it would help to examine what may happen in terms of gravity in something like the beta decay of a neutron in free space? Looking at the intermediate stage you have a large mass gain in the W boson, lasting for a short time ... a student may imagine that this would result in a small gravity pulse.

But of course it does not. And it's not because of the un-reality of the W. It's because gravity couples to energy, not mass, and the energy remains constant.

 

[Simon Bridge]

I wasn't making a definition - I was stating a "rule of thumb". There is a distinction there.
However, if you would like to provide a definition while you are punching holes in other peoples attempts to make sense of things I'm all ears.

What counts as "real" in QM is part of the fun - in a lot of physics, the questions doesn't come up - in a way, the common concepts of "real" don't apply very well ... like the rule of thumb. The Higgs mechanism could be just a clever way of moving numbers around to make predictions ... except we can produce the particle.

I should note that I do not mean to say that virtual particles cannot exist. If they couldn't then there would be no point calling them "virtual" - we'd call them "rubbish" or "nonsense" instead. Perhaps it is my somewhat casual use of the word "real" that is throwing things off?

This is getting close to the point I am making. The virtual particles are the SAME as "real" particles, just with a different mass/momentum.

Yeah - and I'm trying to make a distinction between the models we make and the things we make the models of.

Arizona is real. The Arizona in Things to do and see in Arizona is not - though the one can be used as a guide to the other. Let's not confuse the map for the territory.

But of course it does not. And it's not because of the un-reality of the W.

I don't think I made that claim. I certainly didn't intend to.

It's because gravity couples to energy, not mass, and the energy remains constant.

Which is what I have been saying along with everyone else. I don't think we have any disagreement on that point. I started out thinking in terms of pair production of "real" particles like electron positron pairs.

This is also getting to address what OP is talking about.
Does it matter, in this context, what status we apply to the vacuum stuff - "real" and virtual" have become labels in the model havn't they? There are two ways of answering OPs question: in terms of what we may expect nature to be like if we had a better model, or in terms of what the models say. (And we can do the first one only if we have a good idea about what we've left out of the model we normally use.)

The trouble with gravity is that it takes quite a while to get out of the Newtonian gravitation habits we learned in High School and as an Undergrad.

 

[tom.stoer]

Obviously you can't reach inside a Feynman diagram ... but you can do something very similar. You can select a subgraph of the diagram, and by merely adjusting the momenta of its legs, make it "real".

But that means you transform a propagator into a Hilbert space state!

Assume there are grey, black and white motors, and there are green, red, yellow, ... blue cars. Painting a motor in yellow does not transform the motor into a car ...

 

[Naty1]

BillK: good post...#7

 

[bahamagreen]

"...gravity couples to energy, not mass, and the energy remains constant."

Thx, BillK; that does help.

 

[Haelfix]

In the regime where making the approximations involved in perturbation theory are valid (eg quantum gravity at very large distance scales, where we can talk about virtual particles/gravitons and things like that) then yes, vacuum fluctuations gravitate.

So for instance, we can absolutely say that the Lamb shift of the electron gravitates, and in fact we can make very accurate tests of this using the equivalence principle. You can actually make this conclusion simply by stepping on a weight scale, and noting that over 90% of that value comes from QCD quantum vacuum processes being coupled to gravity.

Are these things 'ontic' or not at the most fundamental levels? No one knows! The rest of this conversation is as usual a matter of semantics and interpretation.

 

[Bill_K]

we can absolutely say that the Lamb shift of the electron gravitates, and in fact we can make very accurate tests of this using the equivalence principle.

Do you have a reference for this? The Lamb shift in Hydrogen is measured to be 4.4 x 10^-6 eV. Since a proton's mass is about 1 GeV, it represents 4.4 x 10^-15 of the Hydrogen atom's mass. But according to Wikipedia, the most accurate test of the equivalence principle was a torsion balance measurement in 1999, with accuracy of 1 part in 5 x 10^-13. So unless I am mistaken, it falls short by two orders of magnitude.

You can actually make this conclusion simply by stepping on a weight scale, and noting that over 90% of that value comes from QCD quantum vacuum processes being coupled to gravity.

I've always heard that 90 percent of the nucleon's mass comes from the kinetic energy of the quarks and gluons and the potential energy between them. These are quite different from vacuum processes, aren't they?

 

[Haelfix]

Do you have a reference for this? The Lamb shift in Hydrogen is measured to be 4.4 x 10^-6 eV. Since a proton's mass is about 1 GeV, it represents 4.4 x 10^-15 of the Hydrogen atom's mass. But according to Wikipedia, the most accurate test of the equivalence principle was a torsion balance measurement in 1999, with accuracy of 1 part in 5 x 10^-13. So unless I am mistaken, it falls short by two orders of magnitude.

Yes, I do, but not for the pure 'Free electron' Lamb shift, but rather the same vacuum diagram for the electron in the 'vicinity' of an Aluminum and Platinum nucleus.

Following an argument by Polchinski, the ratio of gravitational to inertial mass of Aluminum and Platinum is exact to one part in ten to the twelve.
P. Roll, R. Krotkov, and R. Dicke, Ann. Phys. (N. Y.) 26 (1964) 442.
V. Braginsky and V. Panov, Zh. Eksp. Teor. Fiz. 61 (1971) 873.

The electrostatic potential for both is roughly 10^-3 and 3*10^-3 respectively , and therefore we can say that they satisfy the equivalence principle to one part in 10^9.

The 'Lamb shift' loop diagram shifts the electrostatic energy by a factor 10^-3, and therefore we can say that this process does in fact gravitate in the usual way with a precision of one part in 10^6 or so.

I've always heard that 90 percent of the nucleon's mass comes from the kinetic energy of the quarks and gluons and the potential energy between them. These are quite different from vacuum processes, aren't they?

We are going to get into semantics here. The QCD vacuum is nonperturbative, so it depends on how you break things apart and what you call the various components. Suffice it to say that the classical mass and binding energy of quarks and gluons are small relative to an as yet identified quantum component that is about an order of magnitude larger.

Whatever 'that' is, gravitates.

 

[tom.stoer]

That's why it's difficult to talk about virtual particles. Of course the QCD vacuum and the bound states contain something like "quantum fluctuations", not just "pointlike valence quarks"; but these non-perturbative "quantum fluctuations" are different from the usual "virtual particles" in Feynman diagrams.

But as Haelfix says this is semantics; 'it' gravitates.