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

Virtual particles vs. real particles

  1. Feb 10, 2013 #1
    What distiguishes real and virtual particles? Virtual photons, virtual gluons and virtual W particles are often referred to in discussing the interactions they determine. Why and when are they virtual? Why and when are they real?
  2. jcsd
  3. Feb 10, 2013 #2
    No one has even seen/measured/detected a virtual particle. All particles ever detected are real particles. That's a clear difference! :-)

    Virtual particles is a theoretical concept that helps us make mental images about what goes on in interactions. When calculating probabilities for different outcomes in for example particle collisions, quantum field theory gives us very complicated and messy equations to solve. The only practical way to solve them is via perturbation theory, and still just writing down the correct integrals for each term in the expansion is quite tricky math. However, Feynman found some great simplified rules to find the correct integrals by drawing simple pictures of "virtual particles". Finding the right formulas in this way and solving them gives the right answers when we compare to measurement of real colliding particles. So from this comes the concept that the interaction is mediated by exchange of virtual particles. But, again, what anyone sees or detects is always (and will always be) real particles. It is worth to note also that an exact solution to the full quantum field equations would require infinitely many terms in the perturbation expansion, each with more and more virtual particles. And this is just to solve for one probability of two real particles colliding.

    So remembering that virtual particles just is an aid in difficult calculations, they can be useful to "explain" what goes on in the equations in a simplified way. People often say that the force between particles is mediated by virtual particles, but strictly speaking this is just a simplified picture of a more complicated reality. Now, the picture works extremely well in predicting real outcomes, including high energy collisions where several new (real) particles are created. So they are clearly a good picture. I would however still say it is doubtful to claim that virtual particles "exist" since we can never expect to measure one directly.
  4. Feb 10, 2013 #3


    User Avatar
    Science Advisor

    No Dali, you have it exactly backwards. All particles ever detected are virtual particles. A real particle is a solution to the source-free wave equation, created in the infinite past and persisting undisturbed until the infinite future. It's a completely idealized concept. Any particle we detect is a solution to the wave equation with source present. It has had a finite existence, from the point it was created or last interacted, until the point at which we detect it. It is slightly off the mass shell, and therefore virtual.
    Last edited: Feb 10, 2013
  5. Feb 11, 2013 #4
    Hmm, of course strictly speaking you are right. My apologies for confusing the subject, it was a while ago I read my QFT!

    I was thinking that normally the in- and outgoing particles in a Feynman diagram are considered to be on-shell (i.e. "real"), but as you point out that is an approximation since they also have a finite lifetime if they are detected. So the difference between "real" and "virtual" is not as clear-cut as I tried to make it. Still, treating in- and outgoing particles as real (on-shell) is usually a very, very good approximation.
  6. Feb 11, 2013 #5
    Just to clarify aren't the particles being referred to here the so called exchange particles as might appear in QFT? If so what about other particle such as, for example, electrons.According to the statement underlined in post 3 all particles, including electrons are virtual. Is this believed to be the case?
  7. Feb 11, 2013 #6
    The concept of a "real" vs "virtual" particle is not too clearly defined, it should be seen more as a fuzzy term: A "real" particle in the strict sense would perfectly fulfill the on-shell mass condition. As Bill_K wrote, this is never exactly fulfilled, but it can be fulfilled in sufficiently good approximation.
    Feynman writes
    Of course any photon that has a physical effect may be considered as a virtual photon since it is not observed unless it interacts, so that observed photons never really have ω=± k. There are, however, no difficulties in passing to the limit; physically, we know of photons that come from the moon, or the sun, for which the fractional difference between ω and k is very, very small.​
  8. Feb 11, 2013 #7


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    Hey everyone, please put in some consideration to addressing the OP's question AND addressing it at the level that you think he /she might be able to understand. It need not be 100% QFT-accurate all the time.

  9. Feb 11, 2013 #8
    I was thinking the same thing, and I think Dali's first answer was near enough to the question.

    Real particles can be detected, in the normal sense of the idea 'detection', by detectors, from massive CERN detectors to rod and cone photon-detectors in the eye.

    Virtual particles are more detected by balanced Feynman diagram books. Some people have argued that's the only reality they have.
  10. Feb 11, 2013 #9
    According to the wikipedia article on virtual particles:
    Yet, Hawking radiation is caused by the spontaneous creation of virtual particles -- as a consequence of the uncertainty principle -- that are separated by the event horizon and -- I guess -- become "real." Am I missing something?
  11. Feb 11, 2013 #10
    And what's to prevent real particles from hitting a virtual particle (as it is in the process of popping into and out of existence) and then flying off in a different direction?
  12. Feb 12, 2013 #11


    User Avatar
    Science Advisor

    The distinction between real and virtual particles makes sense only on Feynman diagrams. Internal lines of a Feynman diagram describe virtual particles, while external lines describe real particles. That's all!

    Methods in quantum field theory which do not use Feynman diagrams do not use the concept of virtual particles.

    Some popular explanations of Hawking radiation use virtual particles, but such explanations are not based on any actual calculations. The actual calculations of Hawking radiation do not use Feynman diagrams and thus they do not involve virtual particles.
    Last edited: Feb 12, 2013
  13. Feb 12, 2013 #12


    User Avatar
    Science Advisor

    If so, then why real particles are called "real"? What is so real about them?

    A more meaningful answer is that the real particles are the detected ones, but theoretical description of real particles involves an approximation according to which they can be described by the source-free wave equation in infinite past and infinite future.
  14. Feb 12, 2013 #13
    I suppose, then, that if real particles were to be given momentum changes after collisions with virtual particle, then that momentum change would have to be given back so that those virtual particles could pop back out of existence, right?
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook