We estimate the size of the hadronic matrix elements of CP-violating three-gluon and four-gluon Weinberg operators using sum-rule techniques. In the three-gluon case, we are able to reproduce the expressions given in earlier works, while the four-gluon results obtained in this article are new. Our paper therefore represents the first systematic study of contributions to the electric dipole moment of the neutron due to CP-violating dimension-six and dimension-eight operators. We provide many details on both the derivation of the sum rules as well as the analysis of the uncertainties that plague our final predictions.Here G A µν is the QCD field strength tensor,G A µν = 1/2 µνρλ G A ρλ with 0123 = +1 denotes its dual, f ABC are the fully anti-symmetric structure constants of SU(3) and c ABCD denote the colour structures defined in (5.20). Lacking the expertise in LQCD as well as the needed computer resources, we will present estimates of the hadronic matrix elements of the operators in (1.1) using QCD sum-rule techniques. In the case of the dimension-six contribution O 6 such a calculation has already been performed in [40], but the latter publication does not provide details on the actual computation making an independent reevaluation worthwhile. Our determination of the hadronic matrix elements of the dimension-eight term O 8 is instead new. Both results will be used in a companion paper [59], where we derive model-independent bounds on CP-violating Higgs-gluon interactions in BSM scenarios with vanishing or highly suppressed light-quark Yukawa couplings.Our work is organised as follows. After briefly reviewing the basic idea behind the sum-rule determinations of the hadronic matrix elements of O 6 and O 8 , we discuss in Section 3 the phenomenological side of the sum rules. The operator product expansion (OPE) computation of the dimension-six and dimension-eight contributions is described in Section 4 and Section 5, respectively. The matching and the numerical analysis of the sum rules are performed in Section 6. We conclude in Section 7. Technical details are relegated to several appendices.
General idea behind the sum rulesThe central object for the derivation of the sum-rule estimates for the hadronic matrix elements of operators of the type (1.1) is the following correlation function Π(q 2 ) = i d 4 x e iqx Ω T η(x)η(0) Ω EM,O k , (2.1)