We determine one-loop string corrections to Kähler potentials in type IIB orientifold compactifications with either N = 1 or N = 2 supersymmetry, including D-brane moduli, by evaluating string scattering amplitudes.2 One way to do this is to relax momentum conservation [11,12,13,14,15]. We confirm the validity of this prescription in our case in appendix E by considering a 4-point function. In the case of [10], the result was confirmed by a 3-point function of gravitons in [16].3 Notice the twofold use of the term "Kähler" here, because both the compactification manifold and the moduli space are Kähler.
4The moduli dependence of our result for the N = 2 orientifold is in agreement with the one-loop Kähler metric given in [10] for the case of vanishing open string scalars, there inferred from the one-loop correction to the Einstein-Hilbert term. 4 As another check, we explicitly derive the one-loop correction to the prepotential from the corrected Kähler potential. This confirms that our result is consistent with supersymmetry.Let us next summarize the content and organization of this paper. We consider the same three orientifold models as in [20], i.e. the T 4 /Z 2 × T 2 model with N = 2 supersymmetry (in chapter 2), the T 6 /(Z 2 × Z 2 ) model (in chapter 3) and the T 6 /Z ′ 6 model (in chapter 4), both with N = 1 supersymmetry. In section 2.4 we verify for the T 4 /Z 2 × T 2 model that one can straightforwardly reproduce the tree level sigma-model metrics by calculating 2-point functions on the sphere and disk using the Kähler structure adapted vertex operators (doing so, we built on previous work on disk amplitudes, in particular [21,22,23,24,25,26]). We then continue by determining a particular 2-point function at one-loop that allows us to read off the one-loop Kähler potential. We perform several checks on the result. In section 2.6 we verify that the result is consistent with N = 2 supersymmetry by determining the corresponding prepotential. A second check is performed in appendix C, where we calculate five other 2-point functions for the (vector multiplet) scalars of this model and show that the result is consistent with the proposed Kähler potential. Given the fact that the prepotential in N = 2 theories only gets one-loop and non-perturbative corrections, we conclude that our result holds to all orders of perturbation theory in the T 4 /Z 2 × T 2 case. The main result of chapter 2 is given in formula (2.77).We then continue in chapters 3 and 4 to generalize this result to the N = 1 cases of the T 6 /(Z 2 × Z 2 ) and T 6 /Z ′ 6 models. The main results in these chapters can be found in formulas (3.30) and (4.3), respectively.In section 5 we draw some conclusions and, in particular, translate our results to the T-dual picture with D3-and D7-branes.Finally, we relegated some of the technical details to a series of appendices. 4 The case with D9-brane Wilson line moduli was also considered in [19]. Their integral representation for the Kähler metric (given in (2.4) of that paper) is less straightforward to ...