On Kato's square root problem

A priori estimates in context of PDEs (35B45) Linear operator methods in interpolation, moment and extension problems (47A57) Linear accretive operators, dissipative operators, etc. (47B44)

Review text

The authors investigate abstract versions \[ H:= \sum^n_{i,j=1} A_ic_{ij} A_j+ \sum^n_{i=1} (c_iA_i +A_ic_i') +c_0 \] of second-order partial differential operators defined by sectorial forms on a Hilbert space \({\mathcal H}\). The operators \(A_i\) are supposed to be closed and skew-symmetric with a common dense domain \({\mathcal H}_i\) while \(c_{ij}\), \(c_i\) are bounded on \({\mathcal H}\) and such that the real part of \((c_{ij})\) is strictly positive definite. Assuming \(D(L) \subseteq \cap^n_{i,j=1} D(A_iA_j)\), where \(L:= \sum^n_{i=1} A_i^2\) is defined as a form on \({\mathcal H}_1 \times {\mathcal H}_1\) and the boundedness of \(c_{ij}\) on one of the spaces \(D((I+ L)^{\gamma/2})\), \(\gamma\in (0,1)\), equipped with the graph norm, it is proved that \(D((\lambda I+H)^{1/2}) =D((\lambda I+ H^*)^{1/2})= {\mathcal H}_1\) for all large \(\lambda \in\mathbb{R}\).

Recommendations

Cited in

(16)

This page was built for publication: On Kato's square root problem

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1365121)