Assuming two black holes with the same rest mass $m$ collid coming from infinity with velocity $v$ and impact parameter $b$. Lets ignore spin at first. For which values of $v$ and $b$ would these holes scatter or merge?

enter image description here

Using $G=c=m=1$ one unit of length would be half of the Schwarzschild radius from one of the black holes. Therefore for $b<4$ the holes would clearly merge because the event horizons will overlap. But what would happen for greater values of $b$ is there some simple merging condition?

  • $\begingroup$ FYI: The trajectories will not be as you drew them, with straight segments and a bend in the middle. Each BH will follow a smooth hyperbolic path with the common center of mass of the two BH system as one of the hyperbola's foci. (That means that your parameter, $b$ is not defined as simply as what your diagram shows.) $\endgroup$ Commented Jun 25 at 12:25
  • $\begingroup$ Related: physics.stackexchange.com/q/52315/2451 arxiv.org/abs/gr-qc/0201034 $\endgroup$
    – Qmechanic
    Commented Jun 25 at 12:56
  • $\begingroup$ If they are close enough to merge, they are close enough for general relativity and gravitational radiation to be important. The paths would be an inspiral if just close enough, and not be hyperbolic for a near miss. $\endgroup$
    – mmesser314
    Commented Jun 25 at 13:24
  • $\begingroup$ @SolomonSlow I just draw it for clarification the impact parameter b is the seperation of the BHs for t->-inf. $\endgroup$
    – Okarin
    Commented Jun 25 at 13:51
  • $\begingroup$ @mmesser314 What if the starting velocity is high enough? Won't the paths be hyperbolic even though the holes nearly miss each other. Clearly the emitted radiation won't be as high as for mergers but it may be measurable in the future. $\endgroup$
    – Okarin
    Commented Jun 25 at 13:59

1 Answer 1


The critical impact parameter $b_{\rm crit}(v)$ depends on the highly non-linear general relativistic interaction of the two black holes. The only way we currently have to address this question is with numerical relativity simulations on super computers. I.e. there is no simple expression for $b_{\rm crit}(v)$

Some simple things we can easily infer:

  • $b_{\rm crit}(v)$ is a monotonically decreasing function of $v$
  • $b_{\rm crit}(v)$ diverges as $1/v$ for small v.
  • $b_{\rm crit}(1) > 3\sqrt{3}GM/c^2$, where $M=2m$.
  • $\begingroup$ Thanks that helps me a lot. Where did you get the lower bound on the critical impact parameter from? Do you know of any lists where the critical impact parameter was calculated for a couple of velocities? I found this paper where it was calculated for three velocities but a larger list would be nicer. What if we introduce spin. Do you know if this will hold b_crit(v,s) <= b_crit(v,s=0) for all spins s? Sorry for all these questions. $\endgroup$
    – Okarin
    Commented Jun 25 at 23:39

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.