2
$\begingroup$

I am a newbie at tensor notation and I have been told to prove the identity

$$ (\nabla_a\nabla_b - \nabla_b\nabla_a)X^a_{\ \ \ b}=- R^e_{\ \ \ bcd}X^a_{\ \ \ e}+ R^a_{\ \ \ ecd}X^e_{\ \ \ b} $$

I am aware of the definition of the Riemman tensor $(\nabla_a\nabla_b - \nabla_b\nabla_a)X_c= R_{abc}^{\quad d}X_{d}$ and using the metric connection $\nabla_a g_{bc}=0$ I have already shown $(\nabla_a\nabla_b - \nabla_b\nabla_a)X^c= -R_{abd}^{\quad c}X^{d}$ but still I cannot figure out how to obtain the first one... :(

I am working torsion-free and I am aware of the formula that relates the Riemman tensor with the Christoffel symbols.

PS: My guess is to go with $X^a_{\ \ \ b} = v^a w_b$ but not sure if that is the right way to go.

$\endgroup$
0

1 Answer 1

1
$\begingroup$

PS: My guess is to go with $X^a_{\ \ \ b} = v^a w_b$ but not sure if that is the right way to go.

It is. Schematically: $$ R\sim\left[\nabla,\nabla\right] $$ Any derivative worth that name satisfies: $$ \nabla\left(uv\right)=\left(\nabla u\right)v+u\nabla v $$ Covariant derivatives act like, $$ \nabla\sim\partial+\Gamma $$ for contravariant vectors. $$ \nabla\sim\partial-\Gamma $$ for covariant vectors.

Finally, the tangent space at a point is a vector space. Is that enough of a clue?

$\endgroup$
1
  • $\begingroup$ Okay, I think that shed some light, thanks! $\endgroup$ Mar 19, 2021 at 12:45

Your Answer

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

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