Consider ${\cal N}=2$ supersymmetry with $SU(2)$ global symmetry group. Then both supercharges $Q_{ai},\bar{Q}_{\dot{a}\dot{j}}$ transform by 2 dimensional representation of $SU(2)$. Denote $SU(2)_I$ as the aforementioned group.

Now consider flat $R^4$ with rotation $Spin(4)\cong SU(2)_L\times SU(2)_R$. Then supercharges transform like $(2,1,2)\oplus (1,2,2)$ respectively. I guess we mean in Euclidean field theory sense in this part.

Consider $SU(2)_L\times SU(2)_D\cong Spin(4)\subset SU(2)_L\times SU(2)_R\times SU(2)_I$ where $SU(2)_D$ denotes diagonal part of $SU(2)_R\times SU(2)_I$.

Now the book says supercharges transform as $(2,2)\oplus (1,3)\oplus(1,1)$ under $SU(2)_L\times SU(2)_D$.

I will discount the trivial representations. $(2,2)$ should correspond to $Q_{ai}$'s part. It seems that $(1,3)\oplus (1,1)$ is mixture of $\bar{Q}_{\dot{a}\dot{j}}$ as $1/2\otimes 1/2=1\oplus 0$ where $1/2,1,0$ denote total spin.

$\textbf{Q:}$ Is above correct? It seems that I do not have a systematic way to deal with this splitting. What would be a systematic way of dealing with this?

  • $\begingroup$ @Qmechanic Sorry my bad. It should be arxiv.org/abs/hep-th/9408074. It is pg 8 top and bottom paragraphs. $\endgroup$ – user45765 Dec 17 '18 at 20:44
  • 1
    $\begingroup$ pg 8 in PDF reader; pg 7 in text. $\endgroup$ – Qmechanic Dec 17 '18 at 21:04

Yes, OP is right. Ref. 1 is considering topological twisting. Let $G:=SU(2)$ to simplify notation. Let $K:=G_L\times G_R$ be the (double cover of the) group of Euclidean spacetime rotations. Let $G_D\to H:=G_R\times G_I$ be the diagonal embedding $g\mapsto (g,g)$. An irreducible representation of $H$ corresponds to a (possibly reducible) tensor product of representations of $G_D$. Then:

  • The supercharge $Q_{\alpha i}$ transforms in the ${\bf 2}_L\otimes{\bf 1}_R\otimes{\bf 2}_I$ of $G_L\times G_R\times G_I$, and hence in the ${\bf 1}_R\otimes{\bf 2}_I$ of $H$, which corresponds to the ${\bf 1}_D\otimes{\bf 2}_D\cong {\bf 2}_D$ of $G_D$.

  • The supercharge $\bar{Q}_{\dot{\alpha} i}$ transforms in the ${\bf 1}_L\otimes{\bf 2}_R\otimes{\bf 2}_I$ of $G_L\times G_R\times G_I$, and hence in the ${\bf 2}_R\otimes{\bf 2}_I$ of $H$, which corresponds to the ${\bf 2}_D\otimes{\bf 2}_D\cong {\bf 3}_D\oplus {\bf 1}_D$ of $G_D$.


  1. C. Vafa & E. Witten, arXiv:hep-th/9408074; p. 7.
  • $\begingroup$ Was there a systematic treatment of the procedure(decomposition of representations) as described? Say a textbook or a reference book? Thanks. $\endgroup$ – user45765 Dec 17 '18 at 22:16
  • $\begingroup$ I updated the answers with some links. $\endgroup$ – Qmechanic Dec 17 '18 at 22:29

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.