Given the following Young tableaux

enter image description here

for $SU(3)$, how can I deduce that it corresponds to the adjoint representation? I was thinking that the dimension of this representation is 8, as in the case of the adjoint representation. Is this sufficient?

  • $\begingroup$ For SU(3), you know a height -2 tower is an antiquark ($\bar 3$) and a plain box a quark (3), so this is $q\bar q$, the adjoint. Can you now write down the YT for the adjoint of SU(5)? You need not compute the dimensionality. $\endgroup$ – Cosmas Zachos Oct 27 '20 at 19:09
  • $\begingroup$ I think is the same YT, am I wrong? $\endgroup$ – dfgoe55 Oct 27 '20 at 19:53
  • $\begingroup$ The first column is 4-high, the second 1. $\endgroup$ – Cosmas Zachos Oct 27 '20 at 19:57

One can compute the the weights by filling in the numbers 1,2,3 according to the rule for semi-standard tableaux (not decreasing along the rows, strictly increasing down the columns). Each of the eight possible tableaux gives the eigenvalues of $\lambda_3$ (the number of 1's minus the number of 2's) and $\lambda_8$ (number of 1's plus number of 2's minus twice the number of 3's all divided by $\sqrt 3$). If you plot them you will recognise the weight diagram of the octet (adjoint) rep.

  • $\begingroup$ But following another path... if I start from the action of the group on a three-indices tensor: $\psi^{ijk} \rightarrow U^i_l U^j_m U^k_n \psi^{lmn}$ the I can pass to the action of the algebra on the same tensor: $\psi^{ijk} \rightarrow u^i_l \psi^{ljk} + u^j_l \psi^{ilk} + u^k_l \psi^{ijl}$. Now is it possible to prove the the action of the algebra on $\psi$ is the adjoint action? Maybe writing explicitely the form of $\psi$...some hints? $\endgroup$ – dfgoe55 Oct 27 '20 at 19:40
  • $\begingroup$ @dfgoe55 You need to properly symmetrize your tensor first, else what you have lives in $(3,0)\oplus (1,1)\oplus (0,0)$. It so happens that the only irrep of $su(3)$ with dim=8 is the adjoint, but such luck does not necessarily hold in general. The solution of this answer is unambiguous and would work in general. $\endgroup$ – ZeroTheHero Oct 27 '20 at 20:03
  • $\begingroup$ The structrure of the tensor is the following $\psi^{ijk} = \phi^{ijk} - \phi^{kji} + \phi^{jik} - \phi^{kij}$...is it wrong? $\endgroup$ – dfgoe55 Oct 27 '20 at 20:08

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.