1

I was trying to do some exercises to feel more confident about decomposition of reducible representations into direct sum of irreducible representations. The exercise involves decomposition of direct product of 3 $SU(3)$ vectors, and is the following:

$$3 \otimes 3 \otimes 3 = 3 \otimes (6 \oplus \bar{3}),$$

where $6$ represents the symmetric tensor and $\bar{3}$ represents the antisymmetric tensor.

Focusing on the $3 \otimes 6$ part:

$$ 3 \otimes 6 = q^iT^{\{jk\}}, $$

one should completely symmetrize and antisymmetrize the product into:

$$q^iT^{\{jk\}} = S^{\{ijk\}} + A^{[ijk]}.$$

This way we can show that these are the irreducible representations, and they transform within themselves under some $SU(N)$ transformations.

At this point my question is what is the explicit combination of $q^i$ and $T^{\{jk\}}$ that gives us $S^{\{ijk\}}$ and $A^{[ijk]}$?


Maybe it is redundant, but I want to emphasize what I try to obtain. So I show how I understand the first part of this calculation:

$$3 \otimes 3 = 6 \oplus \bar{3}$$ $$T^{ij} = \psi^i \phi^j$$ $$T^{ij} = S^{\{ij\}} + A^{[ij]}$$ $$S^{\{ij\}} = \frac{\psi^i \phi^j + \psi^j \phi^i}{2}$$ $$A^{[ij]} = \frac{\psi^i \phi^j - \psi^j \phi^i}{2} = \epsilon^{ijk}\epsilon_{klm}\psi^l \phi^m$$

Qmechanic
  • 201,751
Kaan Güven
  • 139
  • 8
  • Related: https://physics.stackexchange.com/q/167680/2451 – Qmechanic Jul 29 '22 at 08:55
  • 1
    Related : My answer as former user diracpaul (shown as user82794) here How to get result 3Χ3=6+3¯ for SU(3) irreducible representations?. – Frobenius Jul 29 '22 at 09:24
  • @Qmechanic indeed the answer of Níckolas Alves in the link you shared gives a way to write the product in terms of completely symmetric and antisymmetric tensors, but I don't know how to do that operation. Should I restate my question as "How to completely symmetrize and antisymmetrize a product of a vector and a symmetric rank 2 tensor?" I don't know if there is a formal way to do that. Is this question too mathematical to discuss in this thread? – Kaan Güven Jul 29 '22 at 10:11
  • 1
    $3\otimes 6$ does not decompose into a symmetric and antisymmetric part. You have such a decomposition when you take the tensor product of two identical representations: it makes so sense to talk about symmetry or antisymmetric of objects that are not identical, like for instance a spin-1 state and a spin-1/2 state. – ZeroTheHero Jul 29 '22 at 20:08
  • 1
  • The 3⊗6 part you wrote is wrong: $T^{{ij}}= S^{{ij}}$ cannot get you a fully antisymmetric part, as @ZeroTheHero pointed out above. – Cosmas Zachos Jul 29 '22 at 21:57
  • 1
    Thank you very much for all your answers. I especially liked @Frobenius' exhaustive answer in the post he mentioned. I think the real answer of my question is answered there. – Kaan Güven Jul 30 '22 at 10:24

0 Answers0