Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

What the term virtual particles referred to

  1. Nov 22, 2006 #1
    what the term "virtual particles" referred to

    I wanted to know what the term "virtual particles" referred to. I found many "descriptions" of them on webpages. Fine.

    Then I went to the index of many advanced QM books, and the term "virtual"
    doesn't really show up as prominently as I thought it would.

    Is virtual "slang"? If so, what is the proper term for a virtual object so that I may focus on those pages that describe them?
  2. jcsd
  3. Nov 22, 2006 #2


    User Avatar

    Staff: Mentor

    "Virtual particle" is a commonly-accepted term. However, it doesn't play a role in ordinary quantum mechanics (Schrödinger's equation, etc.). You don't really encounter virtual particles until you get to quantum field theory, e.g. quantum electrodynamics (QED).
  4. Nov 22, 2006 #3
    In QFT, when we do perturbative expansions, there are internal lines in the diagrams (Greens functions) which to not have free ends. At each vertex (where lines join), we integrate over momenta. Although in QFT, particles interact via exchange of other particles (gauge bosons), these particles are not observed and are a manifestation of the expansion. Thus they are termed virtual, as they don't 'exist' as such.
  5. Nov 22, 2006 #4
    Ok - that seems to fit what I have found by searching this forum. I was looking for a precise location where "virtual" was introduced. It sounds like they are "internal lines in the Feynman diagrams which do not have free ends."

    I'll run with that!
  6. Nov 22, 2006 #5
    Now it is not to say that a virtual particle is a particle which does not exist. Now take for example QED. The particles are electrons, positrons and photons. Although all these particles are 'real' (we only infer existence), when say 2 electrons scatter via the exchange of a photon, this photon is virtual in that we do not observe it, but is a results of the perturbation expansion, and we integrate over all the possible value of momentum that the photon could have, i.e. -infinity to infinity.
  7. Nov 23, 2006 #6


    User Avatar
    Science Advisor
    Homework Helper

    Any particle/line in a Feynman diagram which is not on its mass sheet is called "virtual particle".

  8. Nov 23, 2006 #7


    User Avatar
    Science Advisor
    2018 Award

    That is a good definition of the otherwise often abused and misunderstood notion of a "virtual particle".
  9. Nov 23, 2006 #8
    As I have read many times the virtual particle concept is just a product of the perturbation theory in quantum field theory. It’s another word for internal line or propagator, and so its ‘ontological status’ is questionable, that means without perturbation theory there were no virtual particles.

    Also, it is always point out that real photons are internal lines in are larger system, too, so that the distinction between is not so clear.

    OK. But isn’t there a need for them even without/ before QFT?

    Here is my long-winded, probably nonsense idea of virtual particles:

    Take a proton. The proton has a conservative force field, which means that it conserves the ability to do work. In order to do work, it does not need fuel like an engine, it does not have to decay like an atom and it does not have to be accelerated. In order to do work and to accelerate electrons again and again, it simply is a proton with its conservative force field.

    When an atom decays a real photon is emitted and energy is transferred via a wave motion through a field. But this very field, when no waves are sent through it and it is static, does also transmit interaction, as above described. Now, the interaction of a static conservative field is mutual (electron and proton do work on each other), whereas the wave through the field sent off by the decaying atom has a direction.

    But the mutuality of the conservative force is not immediate, no forces at distance at work.
    Instead, static conservative force fields also send something that transports energy (which of course needs time to arrive, although being very fast since sent by light velocity). While being on its way, this something violates energy conservation, since the sender is now not in a lower energy state after sending away energy. Only after the mutual energy transfer has been completed and both senders have received their messages from one another energy conservation is re-established.

    So that’s the need for virtual particles, for virtual energy messengers. By giving up forces at distance, and introducing messengers, we are confronted with a (momentary, also sometimes very long) violation of energy conservation. These messengers do not obey energy conservation, but fortunately energy-time uncertainty explains this violations and allows a non forces-at-distance description.
  10. Nov 23, 2006 #9
    But before Feynman it was Dirac who first coined the term.
    I think I heard he used to described them to a layman as those particles lurking from a big bad dark:smile:vacuum to pop up in the reality due to the nonzero vacuum "zero point" energy.
    Something like that :uhh:
    Last edited: Nov 23, 2006
  11. Nov 23, 2006 #10


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    There is this view of QFD that it's a theory of Feynman diagrams. People have picked up and run with this and produced theories based on this satz. Basically they are combinatorial, looking at permutations and combinations of Feynman diagrams as the elements of reality (!), and of course the category people come and redefine all that and it's Katie bar the Door. Arxiv papers available on request, or people who are into this can respond with whatever they think is apposite.
  12. Nov 24, 2006 #11
    I know nothing of virtual particles and less about QED, but doesn't this you said makes you think a bit more on the fact that it's the interaction between charged particles, or between EM waves and charges, that seems "more real" than photons meant in the sense of travelling particles?
  13. Nov 25, 2006 #12
    Well, photons are certainly no travelling particles. They are quantized energy and momentum of the EM field. Photons are unlocalized/ "smeared out" due to quantum uncertainty, they have no rest frame, and also they are bosonic, i.e. they are all identical, for short they are very strange for us to picture, but they are real inhabitants of our physical universe.
  14. Nov 25, 2006 #13


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    And what is it that photon counters count?

    Does anybody feel that this is just another of those semantic arguments that goes round and round, but can never ever reach a conclusion?
  15. Nov 25, 2006 #14
    You mean something like a "circular argument"?
    Last edited: Nov 25, 2006
  16. Nov 25, 2006 #15


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    More like the old joke about the two guys arguing over the back fence between their properties: "They can never agree because they are arguing from different premises."
  17. Nov 26, 2006 #16


    User Avatar
    Science Advisor

    The idea of virtual particles or states, transcends perturbation theory. And, the notion of virtual particles is best thought of as a metaphor, a very useful one indeed.

    Look at a deuteron, a bound state of a proton and a neutron. Both nucleons can be called virtual. Why? Because they do not have all the properties of free nucleons, that is E*E -M*M -P*P=0 does not hold. Clearly to see either or both nucleons, one must probe with enough energy to dissociate the deuteron. We know the nucleons are there because when we bash the deuteron, we get a neutron and a proton. But, unless we bash we don't know the composition of the deuteron. Over time it's grown to be convenient to call bound particles, virtual particles.

    With an important caveat, that convenience is most useful for resonances in particle physics. For example, there's the famous 3-3 resonance in pion-neucleon scattering, with both spin and isospin of the system equal to 3/2.. The resonance is an "almost bound state", unstable because energetics favor resonance decay -- the energy of the resonant state is some what higher than just a non-interacting state. One can say this resonance is composed of a virtual nucleon and a virtual pion, because the resonance decays into a nucleon-pion state. Whether one does depends on the problem at hand.

    In ordinary QM, the tunneling particles are virtual when inside a high enough potential barrier -- they have the wrong energy-momentum relationship to be a free particle.

    But, of most importance, the idea of a virtual particle is a simple linguistic convenience, a metaphor helpful, for some at least, in working with QM and QFT. The virtual particle notion is not meant to be precise nor rigorous, but merely helpful.

    Reilly Atkinson

    Reilly Atkinson
  18. Nov 29, 2006 #17

    I'm just saying this for the last time, especially for those who compute one Feynmann diagramm to many. There is someting called conservative force field. So forget your Peskin and Schroeder, we talking first year physics course.

    How do two unaccelerated charges (a concervative force field that is, no external forces acting on these charges) communicate? They can't produce photons, pions, whatever constantly out of nothing. There has to be messengers that violate conservation of energy but obey time-energy uncertainty. I call them virtual particles. Period.
  19. Nov 29, 2006 #18
    I especially refer to the 'There has to be messengers'. In what sense does they have to exist. They are certainly never detected. Classical physics does not require quantised fields for charges to interact via a finite propagation speed of the interaction. Look at GR.
  20. Nov 29, 2006 #19
    The holy church might want to learn that Maxwell's theory interacting with currents goes beyond conservative forces (see the Lorentz Dirac equation).

    Perhaps you have ever thought that these radiative losses are very very small, the coupling constant in front of the dissipation term in the lorentz Dirac equation equals (2 \mu q^2 )/(3 c^3) which is roughly around
    10^{-54} in SI units. Moreover, the Lorentz Dirac equation comes from conservation of total energy momentum (and angular momentum if one includes magnetic dipole degrees of freedom). In the classical theory, there are no vacuum fluctuations, so that invalidates your argument below

    Now, about the word virtual as Reilly explains it. Come on guys, do we call planets in the solar system virtual ???? There is no difference between those planets (regarding this aspect) and the nucleons building the deuteron, Einstein's theory is even predicting a dynamical mass for our earth coming from the gravitational field of the sun (in the same way as the nucleons are dressed by gluons I guess).
    We call them virtual because they are useless to include in the state description for the scattering matrix where we ideally measure and prepare free particles.

    Last edited: Nov 29, 2006
  21. Nov 30, 2006 #20
    Just because something is non-localised and probabilistic doesn't mean it can't interact with something else. A photon has a probability of being detected by a detector... that doesn't make it anyless real imo. It certainly doesn't introduce a circular argument, it just makes it more difficult to model the photon's interactions.
    Last edited: Nov 30, 2006
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook