Homogeneous nilmanifolds attached to representations of compact Lie groups (Q1296392)

Starting from a real representation \((J, V)\) of a Clifford algebra \(Cl(\mathfrak z)\) \textit{A. Kaplan} [Trans. Am. Math. Soc. 258, 147-153 (1980; Zbl 0416.35022)] constructed a two-step nilpotent Lie algebra \(\mathfrak n = \mathfrak z \oplus V\) with centre \(\mathfrak z\) and Lie bracket defined on \(V\) by \(\langle[v,w ],z\rangle = \langle J_z v, w\rangle\) for all \(v,w \in V\), \(z \in \mathfrak z\), where \(\langle\;,\;\rangle\) is a natural inner product on \(\mathfrak n\). Two-step nilpotent Lie groups endowed with a left invariant Riemannian metric are often called two-step homogeneous nilmanifolds. The corresponding simply connected Lie group \(N\) is an H-type group endowed with the left invariant metric determined by \(\langle\;,\;\rangle \). In this paper the author studies another interesting subclass of two-step homogeneous nilmanifolds, with a construction similar to that of Kaplan, but starting from a real representation \((\pi, V)\) of a compact Lie algebra \(\mathfrak g\). More precisely, let \(\mathfrak n = \mathfrak g \oplus V\) be the two-step nilpotent Lie algebra with centre \(\mathfrak g\) and Lie bracket defined on \(V\) by \(\langle[v,w ],z\rangle = \langle\pi(z) v, w \rangle\) for all \(v,w \in V\), \(z \in \mathfrak g\), where \(\langle\;,\;\rangle\) is a fixed \(\mathfrak g\)-invariant inner product on \(\mathfrak n\). The simply connected Lie group \(N (\mathfrak g, V)\) with Lie algebra \(\mathfrak n\) is a two-step homogeneous nilmanifold with the left invariant metric determined by \(\langle\;,\;\rangle\). Using a result due to \textit{C. Gordon} [Can. J. Math. 37, 467-487 (1985; Zbl 0554.53035)], the author proves that these homogeneous nilmanifolds are precisely those which are naturally reductive and he studies also some geometrical aspects of these manifolds finding many parallels with \(H\)-type groups. Moreover, he studies the notion of commutativity and weak symmetry within the class of naturally reductive manifolds \(N (\mathfrak g, V)\) obtaining within the above class the first examples of non-weakly symmetric, naturally reductive spaces and new examples of non-commutative naturally reductive spaces. We recall that a commutative space is a connected Riemannian homogeneous space \(M\) whose algebra of all \(I (M)^0\)-invariant differential operators is commutative, where \(I (M)^0\) is the connected component of the full isometry group \(I (M)\) [\textit{J. Berndt, F. Ricci} and \textit{L. Vanhecke}, Generalized Heisenberg groups and Damek-Ricci harmonic spaces, Lect. Notes Math. 1598, Berlin-Heidelberg: Springer-Verlag (1995; Zbl 0818.53067)]. A connected Riemannian manifold \(M\) is said to be weakly symmetric if for any two points \(p,q \in M\) there exists an isometry of \(M\) mapping \(p\) to \(q\) and \(q\) to \(p\). It was proved by \textit{A. Selberg} [J. Indian Math. Soc. (N.S.) 20, 47-87 (1956; Zbl 0072.08201)] that any weakly symmetric space is a commutative space (with respect to \(I (M)\)-invariance, but this coincides with \(I (M)^0\)-invariance for homogeneous nilmanifolds). Then the manifolds \(N (\mathfrak g, V)\) are the first examples of non-weakly symmetric naturally reductive spaces.