How Real Are Virtual Particles?

[Super_Jew]

How "Real" Are Virtual Particles?

Are they a help to particle interaction or are they physically real? How come they are not observable to us? what i want to find out is if they are an artefact or not.I also know that there is a lot of debate to whether virtual particles exist in reality or not.

 

[sas3]

From what I have read about virtual particles, the Lamb shift is a direct measurement of the effect of virtual particles on the hydrogen atom. So yes they do exist.
But I am by no means an expert on this subject. There are other interactions that require virtual particles to exist such as the Coulomb Force, Weak and Strong Nuclear Forces,
and the Casimir effect to name a few. The key to understanding virtual particles is that there existence is limited by an extremely small time and space but they do leave evidence of their existence.
The fact that they leave evidence of their existence has always bothered me because of the law of conservation but I guess the way they get around this is that they leave a +1 effect and a -1 effect that eventually cancel each other out.

 

[xepma]

Virtual particles come into play in perturbation theory, with an emphasis on Feynman diagrams. In quantum field theory it's all about the correlation function which we want to determine:

$$\langle \phi_1(x_1)\cdots\phi_n(x_n)\rangle$$

Unfortunately, in practically all interacting theories we cannot come up with a decent way of solving these correlation functions. What we can do is come up with methods to approximate them. This is called perturbation theory. This is where virtual particles come into play.

The reason is that, in some sense, a correlation function is calculated by summing over all possible ways in which the transition from the initial to final state can occur (note that a correlation function is nothing but that: we start out with an initial state, then something happens in between, and we end up with a final state). This sum is quite difficult to manage, but we can keep track of all different contributions through use of Feynman diagrams. For instance, the emition of a photon by an electron is the sum of diagrams as drawn in this picture: http://www.strings.ph.qmul.ac.uk/%7Ebigdraw/feynman/lots_diagrams-main.jpg . Keep in mind that when we completely specify the initial and final states, then each diagram is in fact nothing but a number! It is the contribution of that specific progress to the overall correlation function.

What we notice in this "sum of possible transitions" is the creation and annihilation of virtual particles. They are virtual, because they can carry any combination of energy and momentum (real particles satisfy: $$E^2 + p^2 = m^2$$).

In the end, virtual particles come into play because we use them to keep track of our approximation techniques. Hence, they are part of our calculational approach to approximate the physical quantities we would like to calculate (which are always written in terms of correlation functions).

If we somehow can come up with a way to perform these calculations of the correlation functions without the need to use approximation techniques then we wouldn't even have to mention the virtual particles. They are mathematical constructions, nothing more.

 

[stevebd1]

In the case of the Casimir effect, virtual particles popping in and out of existence can push two plates together which suggests that work is being done. If the virtual particles obey the Heisenberg principle, $\Delta E \Delta t \ge \hbar/2$ (the virtual particles being less than $\hbar/2$) how is it they can exert force while complying with the first law of thermodynamics?

 

[jnorman]

IMO, virtual particles are about on par with the concept of pilot waves, and while they serve some purpose in modeling, they are fairly nonsensical in practical terms of actually understanding particle interaction - ie, they are so non-intuitive that they tend to confuse discussion more than assist it.

i believe that virtual particles are an artifact of the endless attempt to understand QM via particle interaction, when in reality (again, IMO) we should be working toward understanding QM via field interaction.

 

[malawi_glenn]

As stated in the many threads elsewhere in this topic, the virtual particle are just a "tool" that comes up when one does calculate observables using perturbation theory in quantum field theories. xepma's answer is good on this.

One can not say that since lamb shift comes from higher order correction terms in QED calculations the virtual particles are real, that is to say that integrals exists for "real" etc.

 

[tiny-tim]

Welcome to PF!

Are they a help to particle interaction or are they physically real? How come they are not observable to us? what i want to find out is if they are an artefact or not.I also know that there is a lot of debate to whether virtual particles exist in reality or not.

Hi Super_Jew! Welcome to PF!

As xepma says …

They are mathematical constructions, nothing more.

And, as Fredrik said in another thread , which I recommend you look at …

I consider virtual particles to be nothing more than a graphic representation of the individual terms in a series expansion of a mathematical expression for a probability amplitude.

Virtual particles are maths, not physics … representation, not reality

 

[Dmitry67]

1. QM has serious problems with REALITY and REALISM. So I would be more careful saying this.
2. math, not physics? Very likely physics IS mathematics.

 

[George Jones]

From a quantum field theory that I recently ordered and that seems quite nice:

The correspondence between the integrals that make up the Dyson series and Feynman diagrams is perfectly precise and well-defined. However, it is customary to to further and think of the Feynman diagrams as schematic pictures of physical processes, and here the interpretation acquires a more imaginative character. ... They are, in short, the infamous virtual particles that are so ubiquitous in physicists' discourse. In the final analysis, the only existence they possesses for certain is as picturesque ways of thinking about the ingredients of the integrals in the Dyson series.

 

[tiny-tim]

"Quantum field theory: a Tourist Guide for Mathematicians" by Gerald B. Folland

From a quantum field theory that I recently ordered and that seems quite nice:

Hi George!

actually , you could have got most of it for free at http://books.google.com/books?id=3J...obzzCQ&sa=X&oi=book_result&resnum=1&ct=result

 

[jmatejka]

2. math, not physics? Very likely physics IS mathematics.

Interesting...I would think Physics is the study of the nature of things, perhaps properties of matter and energy more specifically.

Math,(but not the integers), is an invention of man. It is used as the inexact language of the sciences and engineering. (Calculus gives you the answer to nothing, but it often gets you infinitely close to the answer).

You could argue the flight of the ball is parabolic in nature, but I don't think the ball or the universe knows the formula.

 

[George Jones]

Hi George!

actually , you could have got most of it for free at http://books.google.com/books?id=3J...obzzCQ&sa=X&oi=book_result&resnum=1&ct=result

While I find Google Book Search useful for quick look-ups, I don't find it very useful for serious study.

 

[Dmitry67]

Interesting...I would think Physics is the study of the nature of things, perhaps properties of matter and energy more specifically.

Math,(but not the integers), is an invention of man. It is used as the inexact language of the sciences and engineering. (Calculus gives you the answer to nothing, but it often gets you infinitely close to the answer).

You could argue the flight of the ball is parabolic in nature, but I don't think the ball or the universe knows the formula.

It is not my idea, but I really like it:
http://arxiv.org/abs/0704.0646

 

[Super_Jew]

so they are just as real as unicorns? thanks everybody for not judging me for my name.

 

[Count Iblis]

so they are just as real as unicorns? thanks everybody for not judging me for my name.

Yes, in principle (virtual) unicorns will also make a contribution to the amplitude for some state |A> to evolve to some other state |B>.

 

[jtbell]

so they are just as real as unicorns?

What processes can you calculate to ten decimal places, in agreement with experiment, using unicorns? :uhh:

 

[benk99nenm312]

Are they a help to particle interaction or are they physically real? How come they are not observable to us? what i want to find out is if they are an artefact or not.I also know that there is a lot of debate to whether virtual particles exist in reality or not.

This is an argument that I love, and since you clearly ask in the question whether they or real or not, I can give an opinion.

Math is a representation of what happens in the universe. Laws are governed by mathematical descriptions. You can not explain the universe without math. Similarly, you can not explain the math without the universe. Virtual particles have to be real in some sense, whether they are even particles or not is debatable. Without virtual photons, there would be no law of electromagnetism.

Perturbation theory describes them in mathematical expansions, and then the answer gives you what you want to hear. They are a way of saving QM from inaccuracies. And then, physicists contradict their own math by saying that they can't exist.

If QM is right, then they have to exist in some manner, shape or form. That's my opinion, anyway.

 

[tiny-tim]

What processes can you calculate to ten decimal places, in agreement with experiment, using unicorns? :uhh:

well, d'uh … obviously you also need to use bicorns, tricorns, and and so on

… Without virtual photons, there would be no law of electromagnetism.

Perturbation theory describes them in mathematical expansions, and then the answer gives you what you want to hear. They are a way of saving QM from inaccuracies. …

No, without virtual photons, there would be no perturbation theory of the law of electromagnetism.

The law of electromagnetism itself would still make perfect sense.

 

[Count Iblis]

There exist nonvanishing effects that are zero to all orders in perturbation theory. So, they cannot be described in terms of virtual particles. But you can still compute these effects by resumming the divergent tail of the perturbative expansion. If g is the coupling, you can then find that the effect is something like
exp(-c/g).

 

[benk99nenm312]

well, d'uh … obviously you also need to use bicorns, tricorns, and and so on


No, without virtual photons, there would be no perturbation theory of the law of electromagnetism.

The law of electromagnetism itself would still make perfect sense.

What I meant was more along the lines of that we wouldn't have an accurate explanation as to why charged particles attract / repel. Electromagnetic attraction could still be given by coulumb's law and so forth, but the underlying cause of this would be unknown.

 

[lightarrow]

What I meant was more along the lines of that we wouldn't have an accurate explanation as to why charged particles attract / repel. Electromagnetic attraction could still be given by coulumb's law and so forth, but the underlying cause of this would be unknown.

Don't know if you are right or not about virtual particles, but why the "underlying cause" would be known, with vir. par.? Only because we were grown up with the sensorial experience that two distant objects cannot interact if they don't come in direct contact. But with particles we wouldn't have solved anything, IMO; then we should answer the question: "and why do these particles behave that way? Because of the presence of some fields which..." and we would never finish the cycle.

 

[benk99nenm312]

Don't know if you are right or not about virtual particles, but why the "underlying cause" would be known, with vir. par.? Only because we were grown up with the sensorial experience that two distant objects cannot interact if they don't come in direct contact. But with particles we wouldn't have solved anything, IMO; then we should answer the question: "and why do these particles behave that way? Because of the presence of some fields which..." and we would never finish the cycle.

Perturbation theory completes electromagnetism with an explanation involving probability amplitudes and virtual particles and so forth. Virtual particles are not the only thing at hand. If it was just virtual particles acting, we would not have a full theory, because of the presence of fields. You are right. The underlying cause I am referring to is a deeper sense. Einstein showed how space-time is the cause of gravity. There was no need to believe that two distant objects could interact magically. Similarly, virtual particles, while they are not the full story, are a step to giving you a similar picture. Without them, we are one step closer to magic.

 

[Count Iblis]

You can also ask the question if perhaps real particles also do not exist. Suppose that that quantum mechanics is not a fundamental theory. Suppose there exists some fundamental theory in which you can make probabilistic computations. It could be that you have to do some complicated counting of fundamental states which can be done approximately in a clever way by introducing a path integral formalism and a fictitious Lagrangian that takes the form of the Standard Model Largrangian. The fictitious terms being necessary to make sure all configurations are counted once.


Then the correct interpretation of quantum mechanics would be the Copenhagen interpretation. If you do an experiment in which you think you are manipulating/measuring the spin of an electron, what you are really doing is different. But this can then be described as if you are manipulating a fictitious particle which we call the electron.

 

[Super_Jew]

What processes can you calculate to ten decimal places, in agreement with experiment, using unicorns? :uhh:

you can't,they are not real.I've read that a future theory will or may replace virtual particles? i also thought they exist for like example 1/10,000 of a second physically.

 

[Bob S]

Virtual particles normally exist only if they do not violate the Heisenberg uncertainty principle (delta E delta t < Plancks constant). They are certainly necessary in strong central Coulomb fields (vacuum polarization - - Uehling integral). What would happen if a virtual particle pair were created near a black hole event horizon, and only one escaped?

 

[DrChinese]

Virtual particles normally exist only if they do not violate the Heisenberg uncertainty principle (delta E delta t < Plancks constant). They are certainly necessary in strong central Coulomb fields (vacuum polarization - - Uehling integral). What would happen if a virtual particle pair were created near a black hole event horizon, and only one escaped?

You'd have Hawking radiation, I believe...

 

[Super_Jew]

You'd have Hawking radiation, I believe...

so that means we have evidence for the existence of virtual particles?

 

[tiny-tim]

so that means we have evidence for the existence of virtual particles?

Nope! … there is no evidence whatever for Hawking radiation!

… virtual particles? i also thought they exist for like example 1/10,000 of a second physically.

Virtual particles normally exist only if they do not violate the Heisenberg uncertainty principle (∆E ∆t < Plancks constant).

It's well-known (and fairly obvious) that energy-momentum is not conserved at any point on a Feynman diagram if the particles are all real …

there are two opposite conclusions to be drawn from this:

i] at least one of the particles must be virtual (ie, has the "wrong mass"), or

ii] they are all real, and energy is "borrowed" for a short time (the more energy, the shorter the time).

But surely there's no justification for somehow combining these two
opposites, and saying that virtual particles are allowed by HUP "borrowing"?

They are certainly necessary in strong central Coulomb fields (vacuum polarization - - Uehling integral).

Hi Bob S!

Wot's a Uehling integral?

I googled the phrase, and got only three hits … one of which was you, about four weeks ago.

 

[malawi_glenn]

I have to agree with you here Timmy

i) Borrowing energy from HUP makes the particles real

ii) no evidence for hawking, also i have read the derivation, but i see not references to virtual particles there, it is "just" thermal radiation

Nope! … there is no evidence whatever for Hawking radiation!



It's well-known (and fairly obvious) that energy-momentum is not conserved at any point on a Feynman diagram if the particles are all real …

there are two opposite conclusions to be drawn from this:

i] at least one of the particles must be virtual (ie, has the "wrong mass"), or

ii] they are all real, and energy is "borrowed" for a short time (the more energy, the shorter the time).

But surely there's no justification for somehow combining these two
opposites, and saying that virtual particles are allowed by HUP "borrowing"?


Hi Bob S!

Wot's a Uehling integral?

I googled the phrase, and got only three hits … one of which was you, about four weeks ago.

 

[Bob S]

Hi Bob S!

Wot's a Uehling integral?

I googled the phrase, and got only three hits … one of which was you, about four weeks ago.

See E. A. Uehling, Physical Review 48, pg 55 (1935). It is mentioned in Bjorken and Drell Relativistic Quantum Mechanics (McGraw Hill (1964) page 158.
The Uehing integral is used to calculate the penetration of the charge renormalization cloud (vacuum polarization) by charged particles in a strong central Coulomb field. Bjorken mentions primarily l=0 atomic states, but it also applies to l>=1 states. It is relatively small in electronic atoms, but is a large effect in muonic and pionic atoms.

 

[tiny-tim]

Uehling integral

Virtual particles … are certainly necessary in strong central Coulomb fields (vacuum polarization - - Uehling integral)

See E. A. Uehling, Physical Review 48, pg 55 (1935). It is mentioned in Bjorken and Drell Relativistic Quantum Mechanics (McGraw Hill (1964) page 158.
The Uehing integral is used to calculate the penetration of the charge renormalization cloud (vacuum polarization) by charged particles in a strong central Coulomb field. Bjorken mentions primarily l=0 atomic states, but it also applies to l>=1 states. It is relatively small in electronic atoms, but is a large effect in muonic and pionic atoms.

1935? 1964?

It doesn't seem much-loved. ​

Do 21st-century practitioners really regard the reality of virtual particles as necessary for the explanation of vacuum polarisation?

 

[Naty1]

Post 5,6,7 and some others which claim virtual particles are merely mathematics artifices run the risk so many times proved in history...when math proves to be accurate and our preconceived notions prove simplistic, even naive but definitely false...Just because we can't imagine them being real doesn't mean they aren't.

When I studied semiconductor theory many years ago it was necessary to theorize electrons and holes to account for electrial behavior...I don't know whether holes are still necessary, I hope not, but it sure always seem an accounting/mathematical rather than a physical reality...

On the other hand, Einstein refused to believe in his own GR mathematics..so he stuffed in the cosmological constant to keep the universe in a steady state; Einstein also refused to believe in black holes which were also predicted by his own theory...and as a father of quantum theory he believed "God does not place dice" ...and missed reality a third time.

Sometimes experiment teases out new information; sometimes math does...what's "real" and what's not is often not obvious...

 

[Count Iblis]

Students specializing in experimental particle physics often get a dumbed down introduction to perturbation theory and the Feynman rules. If you follow such course then you may get the impression that virtual particles are real.

But if you read a decent book on QFT, e.g. the one by Itzykson and Zuber, you would never think that.

 

[Bob S]

1935? 1964?

It doesn't seem much-loved. ​

Do 21st-century practitioners really regard the reality of virtual particles as necessary for the explanation of vacuum polarisation?

Virtual vacuum polarization particles are not real, but the vacuum polarization concept in treating strong Coulomb fields and charge renormalization is necessary to explain the hyperfine structure of hydrogen atoms (Lamb shift) and measurement of muonic and pionic atoms. Vacuum polarization is a necessary vertex corrention in all but the lowest order of charged particles interacting with the external field.

Please read this recent article:

The Uehling correction to the energy levels in a pionic atom

S. G. Karshenboim, E. Yu. Korzinin, and V. G. Ivanov
Can. J. Phys. 84(2): 107–113 (2006)

Abstract: We consider a correction to energy levels in a pionic atom induced by the Uehling potential, i.e., by a free electron vacuum-polarization loop. The calculation is performed for circular states (l = n–1). The result is obtained in a closed analytic form as a function of Zα and the pion-to-electron mass ratio. Certain asymptotics of the result are also presented.
==================================
If you can get a copy, please forward one to me. Thanks.

 

[peteratcam]

The important point is that particles are not 'real' either.

We have interacting quantum fields, and it is a nightmare to calculate anything: we are partly rescued by some conservation laws (what is conventionally called 'particle number' might be conserved) but still it is impossible to work anything out.

One scheme for doing calculations is to split the QFT action into a quadratic (harmonic) part, and a non-quadratic part. A harmonic oscillator has quantised energy levels, with equally spaced levels. We consider a harmonic oscillator in its 5th excited state as being 5 particles: the point being that the cost of increasing the level, or decreasing the level is independent of which level we are in already and this feature allows us to pretend we have added 'a particle', or taken away 'a particle'. This is what particle means - a harmonic excitation. This also explains away the confusion surrounding 'identical particles'. It is trivially obvious that a harmonic oscillator in its 7th excited state is identical to the same harmonic oscillator in its 7th excited state!

So *if* we commit to doing calculations by splitting the action into a quadratic (non-interacting) and a non-quadratic part (interacting), then we can benefit from the language of particles (which is easy to deal with because harmonic oscillators are easy to deal with) and then hope to deal with the interaction term using perturbation theory. Virtual particles are an essential tool of perturbation theory (a mathematical technique).

Why is the particle picture so useful and ubiquitous? It is the same reason the harmonic oscillator is so useful and ubiquitous: that any sensible potential well is quadratic as a first approximation.

Both particles and virtual particles are features of the necessity to use perturbation theory on a harmonic (ie solvable) problem to work out anything.

I'm sure some people will disagree.
Peter