Cette thèse concerne le problème de trouver une notion naturelle de «courbure scalaire» en géométrie kählérienne généralisée. L'approche utilisée consiste à calculer l'application moment pour l'action du groupe des difféomorphismes hamiltoniens sur l'espace des structures kählériennes généralisées de type symplectique. En effet, il est bien connu que l'application moment pour la restriction de cette action aux structures kählériennes s'identifie à la courbure scalaire riemannienne. On se limite à une certaine classe de structure kählériennes généralisées sur les variétés toriques notée $DGK_{\omega}^{\mathbb{T}}(M)$ que l'on reconnaît comme étant classifiées par la donnée d'une matrice antisymétrique $C$ et d'une fonction réelle strictement convexe $\tau$ (ayant un comportement adéquat au voisinage de la frontière du polytope moment). Ce point de vue rend évident le fait que toute structure kählérienne torique peut être déformée en un élément non kählérien de $DGK_{\omega}^{\mathbb{T}}(M)$, et on note que cette déformation à lieu le long d'une des classes que R. Goto a démontré comme étant libre d'obstruction. On identifie des conditions suffisantes sur une paire $(\tau,C)$ pour qu'elle donne lieu à un élément de $DGK_{\omega}^{\mathbb{T}}(M)$ et on montre qu'en dimension 4, ces conditions sont également nécessaires. Suivant l'adage «l'application moment est la courbure» mentionné ci-haut, des formules pour des notions de «courbure scalaire hermitienne généralisée» et de «courbure scalaire riemannienne généralisée» (en dimension 4) sont obtenues en termes de la fonction $\tau$. Enfin, une expression de la courbure scalaire riemannienne généralisée en termes de la structure bihermitienne sous-jacente est dégagée en dimension 4. Lorsque comparée avec le résultat des physiciens Coimbra et al., notre formule suggère un choix canonique pour le dilaton de leur théorie. / This thesis is about the problem of finding a natural notion of "scalar curvature" in generalized Kähler geometry. The approach taken here is to compute the moment map for the action of the group of hamiltonian diffeomorphisms on the space of generalized Kähler structures of symplectic type. Indeed, it is well known that the moment map for the restriction of this action to the space of ordinary Kähler structures can be naturally identified with the riemannian scalar curvature. We concern ourselves only with a certain class of generalized Kähler structures on toric manifolds which we denote by $DGK_{\omega}^{\mathbb{T}}(M)$ and which we recognize as being classified by the data of an antisymetric matrix $C$ and a real-valued strictly convex functions $\tau$ (exhibiting appropriate behavior on a neighborhood of the boundary of the moment polytope). This viewpoint makes obvious the fact that any toric Kähler structure can be deformed to a non-Kähler element of $DGK_{\omega}^{\mathbb{T}}(M)$, and we note that this deformation happens along one of the classes which were shown by R. Goto to be unobstructed. We identify sufficient conditions on a pair $(\tau,C)$ for it to define an element of $DGK_{\omega}^{\mathbb{T}}(M)$ and we show that in dimension 4, these conditions are also necessary. Following the adage "the moment map is the curvature" mentioned above, formulas for notions of "generalized Hermitian scalar curvature" and "generalized Riemannian scalar curvature" (in dimension 4) are obtained in terms of the function $\tau$. Finally, an expression for the generalized Riemannian scalar curvature in terms of the underlying bi-Hermitian structure is found in dimension 4. When compared with the results of the physicists Coimbra et al., our formula suggests a canonical choice for the dilaton of their theory.

Non-linear sigma models with extended supersymmetry have constrained target space geometries, and can serve as effective tools for investigating and constructing new geometries. Analyzing the geometrical and topological properties of sigma models is necessary to understand the underlying structures of string theory. The most general two-dimensional sigma model with manifest N=(2,2) supersymmetry can be parametrized by chiral, twisted chiral and semichiral superfields. In the research presented in this thesis, N=(4,4) (twisted) supersymmetry is constructed for a semichiral sigma model. It is found that the model can only have additional supersymmetry off-shell if the target space has a dimension larger than four. For four-dimensional target manifolds, supersymmetry can be introduced on-shell, leading to a hyperkähler manifold, or pseudo-supersymmetry can be imposed off-shell, implying a target space which is neutral hyperkähler. Different sigma models and corresponding geometries can be related to each other by T-duality, obtained by gauging isometries of the Lagrangian. The semichiral vector multiplet and the large vector multiplet are needed for gauging isometries mixing semichiral superfields, and chiral and twisted chiral superfields, respectively. We find transformations that close off-shell to a N=(4,4) supersymmetry on the field strengths and gauge potentials of the semichiral vector multiplet, and show that this is not possible for the large vector multiplet. A sigma model parametrized by chiral and twisted chiral superfields can be related to a semichiral sigma model by T-duality. The N=(4,4) supersymmetry transformations of the former model are linear and close off-shell, whereas those of the latter are non-linear and close only on-shell. We show that this discrepancy can be understood from T-duality, and find the origin of the non-linear terms in the transformations.