4
$\begingroup$

This is a question regarding a calculation in perturbative GR. We have :

$g_{\mu\nu} = \eta_{\mu\nu}+h_{\mu\nu}$

where $h_{\mu\nu}$ is a small perturbation around the flat spacetime metric. In linearized theory, we ignore terms which grow as $h^2$ and higher.

Can you point me to a reference which provides the Curvature tensor, Ricci tensor and Ricci scalar to cubic powers in h, and not just the linear term?

$\endgroup$
2
  • $\begingroup$ See the paper below - you might have to do a little extra work but everything you need should be there. $\endgroup$ Commented Jun 24, 2019 at 7:46
  • 1
    $\begingroup$ arxiv.org/abs/hep-th/9411092v1 $\endgroup$ Commented Jun 24, 2019 at 7:46

1 Answer 1

16
$\begingroup$

Expansions around an arbitrary background. You can later put background curvatures to zero to get the expansion around flat space.

Ricci scalar up to $\mathcal{O}(h^3)$:

$$R + \epsilon (- h^{\alpha \beta } R_{\alpha \beta } + \nabla_{\beta }\nabla_{\alpha }h^{\alpha \beta } - \nabla_{\beta }\nabla^{\beta }h^{\alpha }{}_{\alpha }) + \epsilon^2 (2 h_{\alpha }{}^{\gamma } h^{\alpha \beta } R_{\beta \gamma } - h^{\alpha \beta } h^{\gamma \lambda } R_{\alpha \gamma \beta \lambda } + h^{\alpha \beta } \nabla_{\beta }\nabla_{\alpha }h^{\gamma }{}_{\gamma } - \tfrac{1}{4} \nabla_{\beta }h^{\gamma }{}_{\gamma } \nabla^{\beta }h^{\alpha }{}_{\alpha } - \nabla_{\alpha }h^{\alpha \beta } \nabla_{\gamma }h_{\beta }{}^{\gamma } + \nabla^{\beta }h^{\alpha }{}_{\alpha } \nabla_{\gamma }h_{\beta }{}^{\gamma } - 2 h^{\alpha \beta } \nabla_{\gamma }\nabla_{\beta }h_{\alpha }{}^{\gamma } + h^{\alpha \beta } \nabla_{\gamma }\nabla^{\gamma }h_{\alpha \beta } - \tfrac{1}{2} \nabla_{\beta }h_{\alpha \gamma } \nabla^{\gamma }h^{\alpha \beta } + \tfrac{3}{4} \nabla_{\gamma }h_{\alpha \beta } \nabla^{\gamma }h^{\alpha \beta }) + \epsilon^3 (-2 h_{\alpha }{}^{\gamma } h^{\alpha \beta } h_{\beta }{}^{\lambda } R_{\gamma \lambda } + h_{\alpha }{}^{\gamma } h^{\alpha \beta } h^{\lambda \sigma} R_{\beta \lambda \gamma \sigma} - \tfrac{3}{4} h^{\alpha \beta } \nabla_{\alpha }h^{\gamma \lambda } \nabla_{\beta }h_{\gamma \lambda } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\alpha }h^{\gamma }{}_{\gamma } \nabla_{\beta }h^{\lambda }{}_{\lambda } - h^{\alpha \beta } \nabla_{\beta }h^{\lambda }{}_{\lambda } \nabla_{\gamma }h_{\alpha }{}^{\gamma } - h^{\alpha \beta } \nabla_{\beta }h_{\alpha }{}^{\gamma } \nabla_{\gamma }h^{\lambda }{}_{\lambda } - h_{\alpha }{}^{\gamma } h^{\alpha \beta } \nabla_{\gamma }\nabla_{\beta }h^{\lambda }{}_{\lambda } + \tfrac{1}{2} h^{\alpha \beta } \nabla_{\gamma }h^{\lambda }{}_{\lambda } \nabla^{\gamma }h_{\alpha \beta } + h^{\alpha \beta } \nabla_{\gamma }h_{\alpha }{}^{\gamma } \nabla_{\lambda }h_{\beta }{}^{\lambda } + 2 h^{\alpha \beta } \nabla_{\beta }h_{\alpha }{}^{\gamma } \nabla_{\lambda }h_{\gamma }{}^{\lambda } - h^{\alpha \beta } \nabla^{\gamma }h_{\alpha \beta } \nabla_{\lambda }h_{\gamma }{}^{\lambda } + h^{\alpha \beta } h^{\gamma \lambda } \nabla_{\lambda }\nabla_{\beta }h_{\alpha \gamma } - h^{\alpha \beta } h^{\gamma \lambda } \nabla_{\lambda }\nabla_{\gamma }h_{\alpha \beta } + 2 h_{\alpha }{}^{\gamma } h^{\alpha \beta } \nabla_{\lambda }\nabla_{\gamma }h_{\beta }{}^{\lambda } - h_{\alpha }{}^{\gamma } h^{\alpha \beta } \nabla_{\lambda }\nabla^{\lambda }h_{\beta \gamma } + h^{\alpha \beta } \nabla_{\beta }h_{\gamma \lambda } \nabla^{\lambda }h_{\alpha }{}^{\gamma } + \tfrac{1}{2} h^{\alpha \beta } \nabla_{\gamma }h_{\beta \lambda } \nabla^{\lambda }h_{\alpha }{}^{\gamma } - \tfrac{3}{2} h^{\alpha \beta } \nabla_{\lambda }h_{\beta \gamma } \nabla^{\lambda }h_{\alpha }{}^{\gamma })$$

Ricci Tensor up to $\mathcal{O}(h^3)$:

$$ R_{\mu \nu } + \epsilon (\tfrac{1}{2} h_{\nu }{}^{\alpha } R_{\mu \alpha } + \tfrac{1}{2} h_{\mu }{}^{\alpha } R_{\nu \alpha } - h^{\alpha \beta } R_{\mu \alpha \nu \beta } - \tfrac{1}{2} \nabla_{\alpha }\nabla^{\alpha }h_{\mu \nu } + \tfrac{1}{2} \nabla_{\mu }\nabla_{\alpha }h_{\nu }{}^{\alpha } + \tfrac{1}{2} \nabla_{\nu }\nabla_{\alpha }h_{\mu }{}^{\alpha } - \tfrac{1}{2} \nabla_{\nu }\nabla_{\mu }h^{\alpha }{}_{\alpha })+ \epsilon^2 (- \tfrac{1}{2} h^{\beta \gamma } h_{\nu }{}^{\alpha } R_{\mu \beta \alpha \gamma } + h_{\alpha }{}^{\gamma } h^{\alpha \beta } R_{\mu \beta \nu \gamma } - \tfrac{1}{2} h^{\beta \gamma } h_{\mu }{}^{\alpha } R_{\nu \beta \alpha \gamma } - \tfrac{1}{4} \nabla_{\alpha }h^{\beta }{}_{\beta } \nabla^{\alpha }h_{\mu \nu } + \tfrac{1}{2} \nabla^{\alpha }h_{\mu \nu } \nabla_{\beta }h_{\alpha }{}^{\beta } + \tfrac{1}{2} h^{\alpha \beta } \nabla_{\beta }\nabla_{\alpha }h_{\mu \nu } - \tfrac{1}{2} \nabla_{\alpha }h_{\nu \beta } \nabla^{\beta }h_{\mu }{}^{\alpha } + \tfrac{1}{2} \nabla_{\beta }h_{\nu \alpha } \nabla^{\beta }h_{\mu }{}^{\alpha } + \tfrac{1}{4} \nabla_{\alpha }h^{\beta }{}_{\beta } \nabla_{\mu }h_{\nu }{}^{\alpha } - \tfrac{1}{2} \nabla_{\beta }h_{\alpha }{}^{\beta } \nabla_{\mu }h_{\nu }{}^{\alpha } - \tfrac{1}{2} h^{\alpha \beta } \nabla_{\mu }\nabla_{\beta }h_{\nu \alpha } + \tfrac{1}{4} \nabla_{\mu }h^{\alpha \beta } \nabla_{\nu }h_{\alpha \beta } + \tfrac{1}{4} \nabla_{\alpha }h^{\beta }{}_{\beta } \nabla_{\nu }h_{\mu }{}^{\alpha } - \tfrac{1}{2} \nabla_{\beta }h_{\alpha }{}^{\beta } \nabla_{\nu }h_{\mu }{}^{\alpha } - \tfrac{1}{2} h^{\alpha \beta } \nabla_{\nu }\nabla_{\beta }h_{\mu \alpha } + \tfrac{1}{2} h^{\alpha \beta } \nabla_{\nu }\nabla_{\mu }h_{\alpha \beta }) + \epsilon^3 (\tfrac{1}{2} h_{\beta }{}^{\lambda } h^{\beta \gamma } h_{\nu }{}^{\alpha } R_{\mu \gamma \alpha \lambda } - h_{\alpha }{}^{\gamma } h^{\alpha \beta } h_{\beta }{}^{\lambda } R_{\mu \gamma \nu \lambda } + \tfrac{1}{2} h_{\beta }{}^{\lambda } h^{\beta \gamma } h_{\mu }{}^{\alpha } R_{\nu \gamma \alpha \lambda } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\alpha }h_{\mu \nu } \nabla_{\beta }h^{\gamma }{}_{\gamma } - \tfrac{1}{2} h^{\alpha \beta } \nabla_{\alpha }h_{\mu }{}^{\gamma } \nabla_{\beta }h_{\nu \gamma } - \tfrac{1}{2} h^{\alpha \beta } \nabla_{\alpha }h_{\mu \nu } \nabla_{\gamma }h_{\beta }{}^{\gamma } + \tfrac{1}{2} h^{\alpha \beta } \nabla_{\alpha }h_{\mu }{}^{\gamma } \nabla_{\gamma }h_{\nu \beta } - \tfrac{1}{2} h_{\alpha }{}^{\gamma } h^{\alpha \beta } \nabla_{\gamma }\nabla_{\beta }h_{\mu \nu } + \tfrac{1}{2} h^{\alpha \beta } \nabla_{\beta }h_{\nu \gamma } \nabla^{\gamma }h_{\mu \alpha } - \tfrac{1}{2} h^{\alpha \beta } \nabla_{\gamma }h_{\nu \beta } \nabla^{\gamma }h_{\mu \alpha } - \tfrac{1}{2} h^{\alpha \beta } \nabla_{\beta }h_{\alpha \gamma } \nabla^{\gamma }h_{\mu \nu } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\gamma }h_{\alpha \beta } \nabla^{\gamma }h_{\mu \nu } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\beta }h^{\gamma }{}_{\gamma } \nabla_{\mu }h_{\nu \alpha } + \tfrac{1}{2} h^{\alpha \beta } \nabla_{\gamma }h_{\beta }{}^{\gamma } \nabla_{\mu }h_{\nu \alpha } + \tfrac{1}{2} h^{\alpha \beta } \nabla_{\beta }h_{\alpha \gamma } \nabla_{\mu }h_{\nu }{}^{\gamma } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\gamma }h_{\alpha \beta } \nabla_{\mu }h_{\nu }{}^{\gamma } + \tfrac{1}{2} h_{\alpha }{}^{\gamma } h^{\alpha \beta } \nabla_{\mu }\nabla_{\gamma }h_{\nu \beta } - \tfrac{1}{2} h^{\alpha \beta } \nabla_{\mu }h_{\alpha }{}^{\gamma } \nabla_{\nu }h_{\beta \gamma } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\beta }h^{\gamma }{}_{\gamma } \nabla_{\nu }h_{\mu \alpha } + \tfrac{1}{2} h^{\alpha \beta } \nabla_{\gamma }h_{\beta }{}^{\gamma } \nabla_{\nu }h_{\mu \alpha } + \tfrac{1}{2} h^{\alpha \beta } \nabla_{\beta }h_{\alpha \gamma } \nabla_{\nu }h_{\mu }{}^{\gamma } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\gamma }h_{\alpha \beta } \nabla_{\nu }h_{\mu }{}^{\gamma } + \tfrac{1}{2} h_{\alpha }{}^{\gamma } h^{\alpha \beta } \nabla_{\nu }\nabla_{\gamma }h_{\mu \beta } - \tfrac{1}{2} h_{\alpha }{}^{\gamma } h^{\alpha \beta } \nabla_{\nu }\nabla_{\mu }h_{\beta \gamma }) $$

Riemann tensor up to $\mathcal{O}(h^3)$:

$$ R_{\mu \nu \rho \sigma } + \epsilon (\tfrac{1}{2} h_{\sigma }{}^{\alpha } R_{\mu \alpha \nu \rho } - \tfrac{1}{2} h_{\rho }{}^{\alpha } R_{\mu \alpha \nu \sigma } + \tfrac{1}{2} h_{\sigma }{}^{\alpha } R_{\mu \nu \rho \alpha } - \tfrac{1}{2} h_{\rho }{}^{\alpha } R_{\mu \nu \sigma \alpha } - \tfrac{1}{2} h_{\sigma }{}^{\alpha } R_{\mu \rho \nu \alpha } - \tfrac{1}{2} h_{\nu }{}^{\alpha } R_{\mu \rho \sigma \alpha } + \tfrac{1}{2} h_{\rho }{}^{\alpha } R_{\mu \sigma \nu \alpha } + \tfrac{1}{2} h_{\nu }{}^{\alpha } R_{\mu \sigma \rho \alpha } + \tfrac{1}{2} h_{\mu }{}^{\alpha } R_{\nu \rho \sigma \alpha } - \tfrac{1}{2} h_{\mu }{}^{\alpha } R_{\nu \sigma \rho \alpha } - \tfrac{1}{2} \nabla_{\rho }\nabla_{\mu }h_{\nu \sigma } + \tfrac{1}{2} \nabla_{\rho }\nabla_{\nu }h_{\mu \sigma } + \tfrac{1}{2} \nabla_{\sigma }\nabla_{\mu }h_{\nu \rho } - \tfrac{1}{2} \nabla_{\sigma }\nabla_{\nu }h_{\mu \rho }) + \epsilon^2 (- \tfrac{1}{4} \nabla_{\alpha }h_{\nu \sigma } \nabla^{\alpha }h_{\mu \rho } + \tfrac{1}{4} \nabla_{\alpha }h_{\nu \rho } \nabla^{\alpha }h_{\mu \sigma } + \tfrac{1}{4} \nabla^{\alpha }h_{\nu \sigma } \nabla_{\mu }h_{\rho \alpha } - \tfrac{1}{4} \nabla^{\alpha }h_{\nu \rho } \nabla_{\mu }h_{\sigma \alpha } - \tfrac{1}{4} \nabla^{\alpha }h_{\mu \sigma } \nabla_{\nu }h_{\rho \alpha } + \tfrac{1}{4} \nabla_{\mu }h_{\sigma \alpha } \nabla_{\nu }h_{\rho }{}^{\alpha } + \tfrac{1}{4} \nabla^{\alpha }h_{\mu \rho } \nabla_{\nu }h_{\sigma \alpha } - \tfrac{1}{4} \nabla_{\mu }h_{\rho }{}^{\alpha } \nabla_{\nu }h_{\sigma \alpha } + \tfrac{1}{4} \nabla_{\alpha }h_{\nu \sigma } \nabla_{\rho }h_{\mu }{}^{\alpha } - \tfrac{1}{4} \nabla_{\nu }h_{\sigma \alpha } \nabla_{\rho }h_{\mu }{}^{\alpha } - \tfrac{1}{4} \nabla^{\alpha }h_{\mu \sigma } \nabla_{\rho }h_{\nu \alpha } + \tfrac{1}{4} \nabla_{\mu }h_{\sigma \alpha } \nabla_{\rho }h_{\nu }{}^{\alpha } - \tfrac{1}{4} \nabla_{\alpha }h_{\nu \rho } \nabla_{\sigma }h_{\mu }{}^{\alpha } + \tfrac{1}{4} \nabla_{\nu }h_{\rho \alpha } \nabla_{\sigma }h_{\mu }{}^{\alpha } + \tfrac{1}{4} \nabla_{\rho }h_{\nu \alpha } \nabla_{\sigma }h_{\mu }{}^{\alpha } + \tfrac{1}{4} \nabla^{\alpha }h_{\mu \rho } \nabla_{\sigma }h_{\nu \alpha } - \tfrac{1}{4} \nabla_{\rho }h_{\mu }{}^{\alpha } \nabla_{\sigma }h_{\nu \alpha } - \tfrac{1}{4} \nabla_{\mu }h_{\rho \alpha } \nabla_{\sigma }h_{\nu }{}^{\alpha }) + \epsilon^3 (- \tfrac{1}{4} h^{\alpha \beta } \nabla_{\alpha }h_{\mu \sigma } \nabla_{\beta }h_{\nu \rho } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\alpha }h_{\mu \rho } \nabla_{\beta }h_{\nu \sigma } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\alpha }h_{\nu \sigma } \nabla_{\mu }h_{\rho \beta } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\alpha }h_{\nu \rho } \nabla_{\mu }h_{\sigma \beta } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\mu }h_{\sigma \beta } \nabla_{\nu }h_{\rho \alpha } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\alpha }h_{\mu \sigma } \nabla_{\nu }h_{\rho \beta } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\alpha }h_{\mu \rho } \nabla_{\nu }h_{\sigma \beta } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\mu }h_{\rho \alpha } \nabla_{\nu }h_{\sigma \beta } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\beta }h_{\nu \sigma } \nabla_{\rho }h_{\mu \alpha } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\nu }h_{\sigma \beta } \nabla_{\rho }h_{\mu \alpha } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\mu }h_{\sigma \beta } \nabla_{\rho }h_{\nu \alpha } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\alpha }h_{\mu \sigma } \nabla_{\rho }h_{\nu \beta } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\beta }h_{\nu \rho } \nabla_{\sigma }h_{\mu \alpha } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\nu }h_{\rho \beta } \nabla_{\sigma }h_{\mu \alpha } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\rho }h_{\nu \beta } \nabla_{\sigma }h_{\mu \alpha } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\mu }h_{\rho \beta } \nabla_{\sigma }h_{\nu \alpha } - \tfrac{1}{4} h^{\alpha \beta } \nabla_{\alpha }h_{\mu \rho } \nabla_{\sigma }h_{\nu \beta } + \tfrac{1}{4} h^{\alpha \beta } \nabla_{\rho }h_{\mu \alpha } \nabla_{\sigma }h_{\nu \beta }) $$

Computations performed using xAct. You might also find perturbative expansions of curvatures in some PhD theses, though I don't remember coming across one that has them up to the cubic level.

$\endgroup$
8
  • 1
    $\begingroup$ Did the software also write the LaTex code for you? In the linear approximation, one should use the flat spacetime derivative, no longer the covariant/contravariant one. $\endgroup$
    – DanielC
    Commented Jun 24, 2019 at 11:52
  • $\begingroup$ @DanielC There is a command to typeset expressions in TeX format. The expressions above are for an arbitrary background, as mentioned in the beginning. $\endgroup$
    – Avantgarde
    Commented Jun 24, 2019 at 12:27
  • $\begingroup$ When I do the expansion of $R$, I don't get the term $h^{\alpha\beta}h^{\gamma\lambda} R_{\alpha\gamma\beta\lambda}$. Is it easy to see, what I do wrong? DefManifold[M, 4, {\[Alpha], \[Beta], \[Gamma], \[Lambda], \[Sigma], \ \[Rho]}] DefMetric[-1, metric[-\[Alpha], -\[Beta]], cd, {";", "\[Del]"}, PrintAs -> "g"] DefMetricPerturbation[metric, pert, \[Epsilon]] PrintAs[pert] ^= "h" tmp = (Perturbation[RicciScalarcd[], 1] + Perturbation[RicciScalarcd[], 2] + Perturbation[RicciScalarcd[], 3]) // ExpandPerturbation tmp /. pert[LI[n_], __] :> 0 /; n > 1//ToCanonical $\endgroup$
    – p6majo
    Commented Jun 25, 2019 at 19:51
  • $\begingroup$ Thank you for the xAct reference. Very powerful!!! $\endgroup$
    – p6majo
    Commented Jun 25, 2019 at 20:02
  • $\begingroup$ @p6majo You're welcome. I can't read your comment because it's not in mathjax format. Anyway, provided you've coded correctly, you shouldn't be worried. In order to recover 'missing curvatures', you need to commute some covariant derivatives. That's all. $\endgroup$
    – Avantgarde
    Commented Jun 25, 2019 at 20:35

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.