I'm trying to imagine how/why the direction of a spontaneously emitted photon is random, and how that works and reconciles with the conservation of momentum and kinetic energy.

For intuition, I was trying to make a super simple 1-D scenario of one particle emitting one photon in one dimension. So the photon can either be emitted to the left or the right.

(I'm trying to treat this in a classical way for intuition's sake. And because I've no experience in quantum topics.)

enter image description here

I reason that momentum and kinetic energy would be conserved, and since the system has no motion initially, initial momentum and kinetic energy will be zero.

Since it's a linear problem, conservation of angular momentum equation would just devolve into the linear momentum formula, so I ignore it.

The momentum of a photon is $p = \frac{h}{λ}$ and its kinetic energy would be $KE = \frac{hc}{λ} $ (or so I've found), so from conservation of momentum and kinetic energy, I get the system:

$0 = mv + \frac{h}{λ}$

$0 = \frac{1}{2}mv^{2} + \frac{hc}{λ}$

But this looks like it makes no sense to me! The logic of how I built the simple scenario makes sense in my mind, but when I look at this two equation system I make, it doesn't work at all. $h$ and $c$ are constants and the mass of the particle is set before the problem begins. Also, $λ$ is determined by how much energy the particle is shedding by the emission. Meaning the first equation alone solves for the velocity of the particle after the emission, giving $v = \frac{-h}{λm}$. But that implies a single and unique solution, not two possibilities. What's worse, This value of $v$ doesn't even work with the kinetic conservation formula.

So my reasoning must be going down in flames. I don't know how I should think of this scenario. Any suggestions or help?

  • 1
    $\begingroup$ Why is kinetic energy conserved? You are converting internal energy (electron in an excited state) to external energy (photon) $\endgroup$
    – Jon Custer
    Dec 3, 2020 at 21:26
  • $\begingroup$ @JonCuster that makes a lot of sense. And if I consider replacing the + in the momentum equation with a +/-, that’d produce the two possible solutions... I think $\endgroup$ Dec 3, 2020 at 23:29

1 Answer 1


If you are worried about exact conservation of energy and momentum, you really need to do the problem relativistically. With that, the initial 4-momentum is (with $c=1$):

$$ p^{\mu}_0 = (m, 0, 0, 0) $$

If the photon has energy $\hbar\omega$ in the rest frame of the final state atom, then the atom's mass is reduced by that. With a recoil velocity of $\vec v = v\hat z$, the final state 4-momentum is:

$$ p^{mu}_{atom} = \gamma(m-\hbar\omega, 0, 0, (m-\hbar\omega)v)$$

Meanwhile, the photon has 4-momentum:

$$ p^{\mu}_{\gamma} = \sqrt{\frac{1+v}{1-v}}(\hbar\omega, 0, 0, \hbar\omega)$$

where the square root out front accounts for the Doppler shift from the final state atom's rest-frame to the center of mass.

You then solve for everything according to:

$$ p^{\mu}_0 = p^{mu}_{atom} + p^{\mu}_{\gamma} $$

But don't bother, because in an atomic transition:

$$ \hbar\omega \approx 1\,{\rm eV} $$

so the even the lightest recoiling atom is nonrelativsitic with:

$$ K.E. = \frac{p^2}{2m} = \frac 1 2 \frac{(1\,{\rm eV/c})^2}{938\,{\rm MeV/c^2}^2} \approx 5 \times 10^{-10}\,{\rm eV}$$ which is negligible: you can conserve momentum only and get the right answer.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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