Like the title says, are electromagnetic waves only caused by accelerating charges? Are there any other methods for the generation of photons?

If electromagnetic waves are ONLY able to be created by accelerating charges, why is it that when an electron absorbs a photon (and it gets accelerated) it doesn't emit an electromagnetic wave?

  • 1
    $\begingroup$ No, an accelerated magnet also produces electromagnetic radiation $\endgroup$ – Wolphram jonny Oct 1 '19 at 15:36
  • 1
    $\begingroup$ @Wolframjohny I would put that in the category of accelerated charges. $\endgroup$ – my2cts Oct 1 '19 at 15:50
  • $\begingroup$ In principle accelerated magnetic charges would create waves, too. But magnetic monopoles have not been shown to exist, so that's rather speculative. $\endgroup$ – dmckee --- ex-moderator kitten Oct 1 '19 at 17:28
  • 1
    $\begingroup$ @my2cts I disagree sciencedirect.com/science/article/pii/S0927650514001017 $\endgroup$ – Wolphram jonny Oct 1 '19 at 23:16

The classical electromagnetic waves as modeled by Maxwell's equations solutions, are one framework, the classical one.

Electrons and photons are elementary particles and are in the quantum mechanical framework.

The classical framework, emerges smoothly from the quantum framework at the overlap kinematic regions. One cannot use the same mathematics without critical thinking.

The electron does not absorb a photon. A free electron interacts with a photon according to quantum electrodynamic rules, and it is not a classical electrodynamics acceleration or deceleration. An electron bound to matter changes energy levels and the photon is absorbed or emitted from the atom/molecule/lattice. All these interactions are calculable with Feynman diagrams.


For example these are the first order diagrams that contribute to the free electrom photon interaction that can be used to give the probability of interaction and statistical plots of energy and momentum.

How classical fields emerge from quantum is a story that needs the mathematics of quantum field theory.

The classical observation that accelerating charges create electromagnetic waves as well as changing magnetic fields may also do, is beautifully explained with the classical electrodynamics of Maxwell's equations.

| cite | improve this answer | |
  • $\begingroup$ This was a very good answer, and it also cleared up my question. Too bad I can't set two responses as the answer :) $\endgroup$ – Queso Pez Oct 1 '19 at 16:49
  • 1
    $\begingroup$ Anna's answers are always good! They are worth waiting for before picking the winner. lol $\endgroup$ – jmh Oct 1 '19 at 19:05

There are a few ways to create photons:

  1. accelerating charges (like in your example), but based on the comments, you need to count accelerating magnetic dipoles too.


  1. electron relaxation, that is when an excited electron as per QM at a higher energy level moves to a lower energy level, and the difference in energy is emitted as a photon


  1. molecules rotational, vibrational energies can be converted to photon emission.


Emission can happen in two ways:

  1. spontaneous emission, when a QM system transitions from an excited state to a lower energy state and emits a photon


  1. stimulated emission, is when an incoming photon with specific energy can interact with an excited electron, causing it to drop to a lower energy level


Now I understand you are confused because when an electron absorbs a photon, it moves to a higher energy level as per QM. This move is what you imagine as a classical move and acceleration. In reality, this electron does not accelerate, it just moves to a higher energy level as per QM. After this excitation, the electron will relax. This relaxation is when the electron moves back to a lower energy level, and emits a photon. Still, the electron does not accelerate as it moves to a lower energy level, as per QM. It is the difference in energy level (of the electron/atom) that is emitted as a photon. The electron does not accelerate as it moves between the energy levels as per QM, it just starts existing at a different energy level as per QM.

| cite | improve this answer | |
  • 1
    $\begingroup$ why the downvote? $\endgroup$ – Árpád Szendrei Oct 1 '19 at 19:50
  • $\begingroup$ I didnt downvote, but you forgot an accelerating magnetic dipole sciencedirect.com/science/article/pii/S0927650514001017 $\endgroup$ – Wolphram jonny Oct 1 '19 at 23:18
  • $\begingroup$ @Wolphramjonny ok I will edit $\endgroup$ – Árpád Szendrei Oct 2 '19 at 0:36
  • 1
    $\begingroup$ This is a bunch of vaguely related links thrown against the wall with no explanation of the context and no analysis of how they pertain to the question asked (hint: every last one of them—including the one Jonny keeps posting—are arguably related to the acceleration of charge). $\endgroup$ – dmckee --- ex-moderator kitten Oct 2 '19 at 1:41
  • 2
    $\begingroup$ When an electron falls from one shell to a lower one, it's wave function transitions smoothly from one eigenvector to another, creating a superposition of the two eigenvectors with changing amplitudes. This superposition wave function has the property of oscillating precisely with the frequency of the emitted photon. It's the expected location of the electron that oscillates. So, electron relaxation actually falls within the accelerated charge case. $\endgroup$ – cmaster - reinstate monica Oct 2 '19 at 2:16

EM fields are only coupled to charge-current, so propagating fields can only be absorbed or emitted in principle by accelerating charges.

Nevertheless one can distinguish electric and magnetic dipole radiation. Higher multipole radiation also exists. The first describes atomic and molecular transition, electronic or optical active vibrational and rotational. The latter describes ordinary emission of radiation signals (orbital magnetic moment) , the hydrogen 21 cm line, EPR, NMR , MRI.

Nevertheless, the magnetic multipole radiation comes from changing electric current. Current can only change if charge is accelerated. This is even true for time varying spin density , the divergency of which is a time varying current.

| cite | improve this answer | |
  • 1
    $\begingroup$ a time varying magnetic field can be generated by mechanically oscillating a bar magnet that is also electrically an insulator, and will result in EM radiation without any electric charge being accelerated in a macroscopic sense. I think @Wolfram_johnny was talking about this, $\endgroup$ – hyportnex Oct 1 '19 at 20:59
  • $\begingroup$ sciencedirect.com/science/article/pii/S0927650514001017 $\endgroup$ – Wolphram jonny Oct 1 '19 at 23:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.