3
$\begingroup$

Already Delius gave a construction leading from any simple Lie algebra to a quantum Lie algebra $L$. The extension to semisimple seems simple (sorry :-) to me (see below). My question: Can this work for non-semisimple Lie algebras? My guess is "yes, for some subset", for the following reason.

If $L_1$ has the S matrix $S_1$ for irrep $R_1$, likewise $L_2$ $S_2$ $R_2$, then $L_1\bigoplus L_2$ has $S_1\bigotimes S_2$ for irrep $R_1\bigotimes R_2$ (just my hypothesis! - I hope I got the tensor operators right this time - example: choose two copies of $L=A_1$ with the spin $1/2$ irrep of dimension $2$, the S matrix has dimension $2^2$, the tensor product $4^2$, and $4$ is the lowest irrep dimension of $A_1\bigoplus A_1$, and it clebsches as $4\bigotimes 4=9+(3+3)_{ad}+1$.)

But now you can also "add" any two S matrices as $S_1 "\bigoplus" S_2$ (this is not exactly right, you must afterwards fill up all "cross" entries $12|21$ with an "1" to get the Yang-Baxter equation working - I am too lazy to actually prove it works -, but in any case the dimensions of the S matrices just add). Again, "add" the S matrices of two copies of $2$ of $A_1$, the S matrix has dimension $2^2$, the tensor product $4^2$, but it now clebsches as $4\bigotimes 4=4+4+(3+3)_{ad?}+1+1$. Already the two $1$ smell fishy, I guess the reduction is incomplete, in any case the result is something completely different than that of $A_1\bigoplus A_1$.

I've been told on SO that $\bigoplus$ is the only operation you can apply to any two Lie algebras. So two main possibilities exist - is one of them correct?

  • Just because $S$ is a valid S matrix, it doesn't mean that it is the S matrix of some $L$. Question: Do any valid $S$ corresponds to some $L$? Do any $L$ (not only semisimple) have a valid $S$? (I guess this corresponds to my first question.)
  • Some other operation $\bigodot$ can be applied to any two members of some subset of Lie algebras, the set generated under $\bigodot,\bigoplus$ is closed, and $S_1"\bigoplus"S_2$ indeed is the S matrix of $L_1\bigodot L_2$.
$\endgroup$

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.