Cartan matrix

The integer matrix encoding the simple-root geometry of a semisimple Lie algebra.
Cartan matrix

Let g\mathfrak{g} be a complex , choose a h\mathfrak{h}, and let Φh\Phi\subset \mathfrak{h}^* be the associated . Fix a set of Δ={α1,,α}\Delta=\{\alpha_1,\dots,\alpha_\ell\}.

Choose any WW-invariant inner product on the real span of Φ\Phi (for example, the one induced by the ).

Definition. The Cartan matrix A=(aij)1i,jA=(a_{ij})_{1\le i,j\le \ell} is defined by

aij  =  2(αi,αj)(αj,αj). a_{ij} \;=\; 2\,\frac{(\alpha_i,\alpha_j)}{(\alpha_j,\alpha_j)}.

Basic properties.

  • aii=2a_{ii}=2 for all ii.
  • For iji\ne j, one has aijZ0a_{ij}\in \mathbb{Z}_{\le 0}, and aij=0a_{ij}=0 iff αi\alpha_i is orthogonal to αj\alpha_j.
  • The matrix AA determines the (and conversely): the off-diagonal entries record the angles and relative lengths between simple roots.

Context. The Cartan matrix is the combinatorial input for the : connected Dynkin diagrams (or indecomposable Cartan matrices) correspond to simple Lie algebras, while disjoint unions correspond to direct sums.