Construction of blow-up solutions for the focusing critical-energy nonlinear wave equation in $\mathbb{R}^4$ and $\mathbb{R}^5$

Abstract: The goal of this paper is to exhibit solutions $u(x,t)$ of the focusing, critical energy, nonlinear wave equation \begin{equation} \partial_{tt}u - \Delta u - |u|^{p-1}u = 0, \quad t \geq 0, \ x \in \mathbb{R}^d, \ d \geq 3, \ p = (d+2)/(d-2) \end{equation} in dimension $d \in {4,5}$ where a finite-time Type II blow-up (meaning the solution blows-up while the energy remains bounded) occurs exactly at $x = t = 0$ with a prescribed polynomial blow-up rate $t^{-1-\nu}$ with $\nu > 1$ when $d = 4$ and $\nu > 3$ when $d = 5$. Such solutions have been constructed by Krieger-Schlag-Tataru in $d = 3$ and by Jendrej in $d = 5$. The result from Jendrej covers the extremal case $\nu = 3$, which we are missing, and any $\nu > 8$. Surprisingly, no one else has treated the $d = 4$ case yet. The major difference between dimensions 4 and 5 in our paper lies in this renormalization procedure. In $d = 4$, we essentially follow Krieger-Schlag-Tataru scheme used for the 3D equation. This scheme has been used with success in other equations such as the 3D-critical NLS, Schr\"odinger maps or wave maps. In all of these cases, the non-linearity is a polynomial, which allows to treat the error terms using simple algebra rules. In $d = 5$, the setup needs to be modified because the nonlinearity of the equation has a much lower regularity and it is not clear anymore how one should deal with the nonlinear error terms.