Local \(\zeta\)-function techniques vs. point-splitting procedure: A few rigorous results (Q1299955)

The author discusses some general properties of local \(\zeta\)-function procedures to renormalize several quantities in \(D\)-dimensional Euclidean quantum field theory in a curved background, for positive scalar operators \(-\Delta+V(x)\), in closed \(D\)-dimensional manifolds. A few considerations are made also for the case of non-closed manifolds. The author performs a general comparison with respect to the more familiar method of point-splitting, concerning the effective Lagrangian and the field fluctuations. It is proven in the paper that, for \(D>1\), both the local \(\zeta\)-function and the point-splitting approaches lead essentially to the same results, apart from some differences in the subtraction procedure of the Hadamard divergences. Since these results are local ones, the author conjectures that this agreement might well also hold when dropping the hypothesis of the manifold being a compact one without boundary. The author proves, in particular, that the local \(\zeta\)-function technique is rigorously founded and produces essentially the same results as the point-splitting procedure, at least concerning the effective Lagrangian and the field fluctuations. He shows that this statement holds for any dimension \(D>1\) in closed manifolds for Friedrichs extensions of a Schrödinger-like real positive smooth operator. He finds, in particular, that with the \(\zeta\)-function procedure one picks out a particular term, \(w_0(x,y)\), in the Hadamard expansion. The presence of a nontrivial kernel of the operator \(-\Delta+V(x)\) may produce however significant differences between the two approaches: the local \(\zeta\)-function method can be still successfully employed, whereas the point-splitting procedure is not completely well-defined in such case. The equivalenee of the local \(\zeta\)-function method and the point-splitting one for treating the one-loop stress tensor has been investigated by the author in a subsequent publication [J. Math. Phys. 40, No. 8, 3843-3875 (1999)].