# Stress tensor of complex scalar field

The canonical stress tensor is defined by

$$T^{\mu\nu}=\frac{\partial\mathcal{L}}{\partial(\partial_{\mu}\phi_a)}\partial^{\nu}\phi_a-\eta^{\mu\nu}\mathcal{L}.$$

where $$\phi_a$$ are the different fields in the theory. For the complex scalar field, the Lagrangian is

$$\mathcal{L}=\partial_{\mu}\phi^{\dagger}\partial^{\mu}\phi-m^2\phi^{\dagger}\phi$$

from which I calculate (treating $$\phi$$ and $$\phi^{\dagger}$$ as two different $$\phi_a$$)

$$T^{\mu\nu}=\partial^{\mu}\phi^{\dagger}\partial^{\nu}\phi+\partial^{\nu}\phi^{\dagger}\partial^{\mu}\phi-\eta^{\mu\nu}\mathcal{L}.$$

However, my book states the answer as

$$T^{\mu\nu}=2\partial^{\mu}\phi^{\dagger}\partial^{\nu}\phi-\eta^{\mu\nu}\mathcal{L}.$$

Is it obvious that this simplification is allowed (or is my calculation wrong)? To me it seems like $$\partial^{\mu}\phi^{\dagger}\partial^{\nu}\phi$$ could be complex, but I can't think of a simple example where it is.

• In the last expression, where is the symmetry b/w $\mu$ and $\nu$ indices?
– KP99
Dec 8, 2021 at 14:42
• Try polar-decomposing $\phi = |\phi| e^{i \Phi}$ and see if the term $\partial^{\mu} \phi^{\dagger} \partial^{\nu} \phi$ is real. Dec 8, 2021 at 15:17
• It is page 77 in Quantum Fields: From the Hubble to the Planck Scale. It says that $T^{\mu\nu}$ is not guaranteed to be symmetric from the definition, but that the result for the complex scalar field is symmetric. Dec 8, 2021 at 15:54
• Polar decomposition gives imaginary cross terms which do not cancel out so I think the book must be wrong. Dec 8, 2021 at 16:32
• Yes, the book is wrong. Author's homepage: web.phys.ntnu.no/~mika/QF.html Dec 8, 2021 at 17:15

1. For a complex scalar field, the canonical stress-energy-momentum (SEM) tensor reads \begin{align}\mp T^{\mu}{}_{\nu}~=~&\frac{\partial {\cal L}}{\partial {\rm Re}\phi_{\mu}}{\rm Re}\phi_{\nu}+\frac{\partial {\cal L}}{\partial {\rm Im}\phi_{\mu}}{\rm Im}\phi_{\nu} -\delta^{\mu}_{\nu}{\cal L}\cr ~=~&\frac{\partial {\cal L}}{\partial \phi_{\mu}}\phi_{\nu}+\frac{\partial {\cal L}}{\partial \phi^{\ast}_{\mu}}\phi^{\ast}_{\nu} -\delta^{\mu}_{\nu}{\cal L}\end{align} for Minkowski sign convention $$(\mp,\pm,\pm,\pm)$$, respectively.
2. It follows that the canonical SEM tensor $$T^{\mu\nu}=T^{\nu\mu}$$ is symmetric for a complex scalar field.