I read the following statement in one of Penrose's paper

zero rest-mass field equations can, with suitable interpretations, be regarded as being conformally invariant.

I take this to imply that if I would like to describe massless scalar fields (for example) in curved spacetimes I should couple them conformally. More precisely, the curved space generalization of the action is $$ - \phi \partial^2 \phi \to - \phi \left( \nabla^2 + \frac{1}{6} R \right) \phi ~~~(d=4) $$ instead of the naive $\partial^2 \to \nabla^2$. Why is this the case?

EDIT: I believe this holds for massive scalar fields as well, though we no longer have conformal invariance.

  • $\begingroup$ Are you suggesting that one needs to include the $R/6$ for purposes of anomaly cancellation in a corresponding QFT? $\endgroup$ – joshphysics Aug 16 '13 at 1:53
  • $\begingroup$ I am not suggesting that. I do not know the origin of adding in the $R/6$. But what I've read is "In the framework of canonical quantum field theory in curved spacetime, the action must necessarily be of the form [mentioned above]". I do not understand why this should necessarily be the case. If there is some anomaly cancellation you are referring to, could you be more specific? $\endgroup$ – Prahar Aug 16 '13 at 1:58
  • $\begingroup$ Prahar when answering to someone's comment you should use @ (user name). e.g. @joshphysics $\endgroup$ – user10001 Aug 16 '13 at 2:27
  • $\begingroup$ This may be helpful: arxiv.org/abs/gr-qc/0205066 'Only for ξ = 1/6 (so-called “conformal coupling”) does the flat space limit of the stress-energy tensor for the scalar field yield an expression with good renormalization properties.' $\endgroup$ – Ben Crowell Aug 16 '13 at 2:46
  • $\begingroup$ @Prahar I have no idea if this has anything to do with anomalies, but whenever I think of conformal invariance and I see the Ricci scalar sitting around, I think of the conformal anomaly...en.wikipedia.org/wiki/Conformal_anomaly $\endgroup$ – joshphysics Aug 16 '13 at 3:22

With the choice of $R/6$ term in four dimensions, the action is invariant under conformal transformations of the metric $g_{\mu\nu} \rightarrow \Omega(x)g_{\mu\nu}$. Such actions are of interest, because if the spacetime itself is conformally flat (for example, spatially flat FRW metric), then the action is equivalent to that of a field in flat spacetime! This is of course a huge simplification and implies that the scalar field is decoupled from gravity.

A detailed derivation may be found in chapter 3 of the book by Birrell and Davies. In $n$ dimensions, the corresponding factor in front of $R$ is $(1/4)(n-2)/(n-1)$.

One could also not put the factor and the study the consequent theory. Conformal invariance action is of course not mandatory but for reasons explained above, nice nonetheless.

  • 1
    $\begingroup$ It should be noted that the above reasoning is classical. It has nothing to do with conformal anomaly, which is a quantum effect. Conformal invariance of the action implies that the trace of the energy-momentum tensor vanishes, but the vacuum expectation of the trace of the EMT maybe non-zero. This is the conformal anomaly, or trace anomaly. $\endgroup$ – Aditya Bawane Aug 16 '13 at 19:24
  • $\begingroup$ I see that. However, I am asking why is this conformal invariance imposed upon us by the requirement of renormalizability? $\endgroup$ – Prahar Aug 16 '13 at 19:35
  • 1
    $\begingroup$ @AdityaBawane: Welcome to physics.SE! If you like, you can edit your own answers rather than posting additional material as a separate comment. $\endgroup$ – Ben Crowell Aug 16 '13 at 21:21

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.