There is a 6 dimensional multiplet belonging to an irreducible representation of a unitary group of rank less than 3. How does one check if the states $|i\rangle$ belong to spin 5/2 representation of $SU(2)$ or they belong to 6 dimensional representation of $SU(3)$?

For a given choice of basis states of the Hilbert space, one can, in principle, obtain a representation of any group, the dimension of the representation being the same as that of the Hilbert space. Just by knowing the basis states, I think it may not be possible to identify which group has been represented. Is my conclusion correct, or there is some point I am missing?


If I know the explicit form of the states I would first apply the SU(2) generators to them. For example apply the Cartan generator $L_3$ and see whether you get the eigenvalues $-5/2, -3/2,...,+5/2$.

  • $\begingroup$ Is the form the Cartan generators known unless the representation itself is known? I was confused regarding if the the Cartan generators remain invariant accross all the equivalent representations of the same group? $\endgroup$ – vnd Oct 29 '14 at 16:33
  • $\begingroup$ Well the explicit form of the algebra generators depend upon a chosen basis. They change their explicit matrix form under similarity transformation. So... what I told you to do will not work unless you're using the same basis for generators and states. Anyway, you definitely need some extra information do determine the algebra. If I give you only six states $|i\rangle$ and nothing else, you can't even say they belong to an irreducible representation. $\endgroup$ – Diracology Oct 29 '14 at 16:43
  • $\begingroup$ So we basically assume first that the states do indeed give a representation of SU(2) and check for eigenvalues with L_3. If not then we assume they give a representation of SU(3) and check eigenvalues with H_1 and H_2. $\endgroup$ – vnd Oct 29 '14 at 16:55
  • $\begingroup$ That's what I would try first. Although it might happen you don't get something useful because the states and the matrices are in different basis. $\endgroup$ – Diracology Oct 29 '14 at 16:59
  • $\begingroup$ Even in the finite group case, the same choice of basis states can give representations of identical dimensions for two non isomorphic groups. For example the binary basis (xy,xz,yz) gives 3 dimensional representations of the Tehtrahedral(T) as well as the Octahedral(O) groups. It was this point that was causing confusion. Thanks for the clarifications. $\endgroup$ – vnd Oct 29 '14 at 17:10

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.