Suppose I have a thin spherically symmetric shell of matter of (rest) mass $M$ that collapses to form a black hole, and later a second thin spherical shell of mass (rest) $M'$, concentric to the first, that collapses too. I have two questions about this process:

  1. Why the radius of the innermost shell 1 cannot be equal to $2M$ when it crosses its horizon? (I know that it cannot be equal to 2M because I have been told so, but I do not know why and, most importantly, I do not know if it is more or less than $2M$).

  2. How does the situation change if the second shell does NOT collapse?


I) For a single$^1$ spherically symmetric thin-shell collapse with (rest) mass $m$, the geometry inside the shell is flat Minkowski space, cf. e.g. this Phys.SE post. Outside the shell the geometry is Schwarzschild geometry with Schwarzschild radius $R_s=2E$ in natural units. Here

$$E~=~ m\sqrt{1+\left(\frac{dR}{d\tau}\right)^2} - \frac{m^2}{2R} \tag{3.70} $$

is the total energy of the thin shell, including the gravitational binding-energy. And here $\tau$ denotes the proper time of the shell. Note that the total energy $E$ is the gravitational mass of the shell, and a constant of motion.

II) If the formula (3.70) should be valid at $R=\infty$, we must demand that the total energy $E>m$ is bigger than the (rest) mass. In that case the Schwarzschild radius $R_s=2E$ cannot be $2m$, cf. OP's question (v2).

III) To reflect the collapse, the sign of the velocity is chosen negative

$$ \frac{dR}{d\tau}~=~-\sqrt{\left(e+\frac{m}{2R}\right)^2-1}~<~0, \qquad e~:=~\frac{E}{m}~>~1. $$

Let us (for the fun of it) integrate to find the elapsed proper time $\Delta\tau$ between two radial positions $R_i>R_f$:

$$ \Delta\tau~=~\tau_f-\tau_i~=~ \int_{R_f}^{R_i} \frac{dR}{\sqrt{\left(e+\frac{m}{2R}\right)^2-1}} ~=~ \int_{R_f}^{R_i} \frac{R~dR}{\sqrt{\left(Re+\frac{m}{2}\right)^2-R^2}} $$ $$~=~\left[ \frac{\sqrt{\left(Re+\frac{m}{2}\right)^2-R^2} }{e^2-1} \right. -\left. \frac{em\ln\left\{2\sqrt{\left(Re+\frac{m}{2}\right)^2-R^2} + 2R\sqrt{e^2-1}+\frac{em}{\sqrt{e^2-1}} \right\}}{2(e^2-1)^{\frac{3}{2}}}\right]_{R=R_f}^{R=R_i} $$


  1. E. Poisson, A Relativist's toolkit, 2004; Subsection 3.9, eq. (3.70).


$^1$ We leave the case of two concentric spherically symmetric thin shells as an exercise for the reader. Obviously, Birkhoff's Theorem comes in handy.

| cite | improve this answer | |

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.