According to Visser M. in the book "Lorentzian Wormholes: from Einstein to Hawking", a realistic model for the stress-energy tensor of the Casimir effect is presented in equation (12.31): $$ T^{\mu \nu} = \sigma \hat{t}^{\mu} \hat{t}^{\nu} [ \delta (z) + \delta(z-a))] + \Theta (z) \Theta (a - z) \frac{\pi^2}{720} \frac{\hbar}{a^4} \left[ \eta^{\mu \nu} - 4 \hat{z}^{\mu} \hat{z}^{\nu} \right],\tag{12.31}$$ Where only the $z$ direction is considered, $\hat{t}$ is the unit vector in the time direction and $a$ is the distance between the plates. Also, $\sigma$ is the mass density of the plates. How can I derive that tensor?

  • $\begingroup$ The second term, between the two plates, is more commonly written $\frac{\pi^2}{720}\frac{\hbar}{a^4}\text{diag}(-1, 1, 1, -3)$. (for example, see the page numbered 30 in this paper). Do you know how to derive that? $\endgroup$
    – Ghoster
    Jun 17, 2023 at 4:58
  • $\begingroup$ @Ghoster I do know how to derive that. But my question is more focused on the first term $\endgroup$ Jun 17, 2023 at 12:55
  • $\begingroup$ First term is just the energy density of the two plates, which are concentrated at z and z-a, therefore the Delta function terms $\endgroup$
    – KP99
    Jun 17, 2023 at 13:03
  • $\begingroup$ @KP99 I'm sorry but I still don't understand why the delta and $\Theta$ function appears $\endgroup$ Jun 18, 2023 at 3:39

1 Answer 1


Delta functions are different from zero only in a point, while Heaviside $H(x)$ step functions are non-zero only at $x>=0$.

So the first term is the positive energy density of the two plates, since the plates are located at z=0 and z=a.

The second term is the negative energy density between the plates, since the product of step function is different from zero only when $0 <= z <= a$.


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.