Analysis of micropolar fluids: existence of potential microflow solutions, nearby global well-posedness, and asymptotic stability (Q829614)

The authors prove existence results for the motion of micropolar fluids. They write the balance equations as: \(\operatorname{div} u=0\), \(\rho (\partial_{t}+u\cdot \nabla)u-\operatorname{div} S=f\), \(j(\partial_{t}+u\cdot \nabla)u- \operatorname{div}C+\kappa (2\omega - \operatorname{curl} u)=g\), in \(\mathbb{R}^{+}\times \mathbb{T}^{3}\), where \(S,C\in \mathbb{R}^{3\times 3}\) are the stress and stress-couple tensors, which involve the vector fields \(u,\omega :\mathbb{R}^{+}\times \mathbb{T}^{3}\rightarrow \mathbb{R}^{3}\), the stress tensor \(S\) having an antisymmetric term, \(\rho ,j\in \mathbb{R}^{+}\) are the fluid density and microrotational inertia, \(\kappa \in \mathbb{R}^{+}\) is the asymmetric viscosity coefficient, and \(f,g\) represent the applied force and microtorque. The initial condition \((u(0),\omega (0))=(u_{0},\omega_{0})\) on \(\mathbb{T}^{3}\) is added. The first main result proves that for any conservative initial configuration for the angular velocity of the microstructure \(\zeta_{0}:\mathbb{T}^{3}\rightarrow \mathbb{R}^{3}\), there exists a microangular velocity \(\zeta :\mathbb{R}^{+}\times \mathbb{T}^{3}\rightarrow \mathbb{R}^{3}\) such that the triple \((u,\omega ,p)=(0,\zeta ,0)\) is the unique solution to the above problem with initial data \((u_{0},\omega_{0})=(0,\zeta_{0})\) and vanishing force and microtorque \(f=g=0\). These potential microflows are smooth in \(\mathbb{R}^{+}\times \mathbb{T}^{3}\) and decay to zero as \(t\rightarrow \infty \). For the proof, the authors introduce the Leray projector to decouple the problem into a pair of vectorial heat equations, and the \(\alpha \)-heat flow mapping \(S_{\alpha }:\overset{o}{H^{r}}(\mathbb{T}^{3};\mathbb{R}^{3})\rightarrow L_{\mathrm{loc}}^{1}(\mathbb{R}^{+};\overset{o}{H^{r}}(\mathbb{T}^{3};\mathbb{R}^{3})) \) defined through \(S_{\alpha }(g)(t)=\sum_{k\in \mathbb{Z}^{3}}\exp (-4\pi^{2}\alpha \left\vert k\right\vert^{2}t)\widehat{g}(k)e_{k}\). They prove properties of these operator and mapping, in connection with spatial and space-time Sobolev spaces. They introduce the time-dependent Lamé system: \(\partial_{t}\omega -(\alpha +\gamma)\Delta \omega -(\beta -\gamma)\nabla \operatorname{div}\omega +\delta \omega =f\), posed in \((0,T)\times \mathbb{R}^{3}\) with the initial condition \(\omega (0)=\omega_{0}\) and the mapping \(S_{\mu ,\nu }(g)(t)=\exp (-\nu t)\sum_{k\in \mathbb{Z}^{3}}\exp (-4\pi^{2}\mu \left\vert k\right\vert^{2}t)\widehat{g}(k)e_{k}\). They finally introduce the notion of potential microflow and they linearize the above problem around such potential microflow. The existence result is proved applying abstract functional arguments. The second main result proves that if \(\zeta_{0}:\mathbb{T}^{3}\rightarrow \mathbb{R}^{3}\) is a sufficiently regular conservative generator for a potential microflow \(\zeta \) as proved in the previous result, the problem is globally well-posed near \(\zeta_{0}\) in the sense that to each tuple of data/forcing/microtorquing \((u_{0},\omega_{0},f,g)\) belonging to an open subset of an appropriate function space that contains \((0,\zeta_{0},0,0)\), there exists a unique global-in-time solution \((u,\omega ,p)\) to the problem that belongs to another appropriate function space. Moreover, the map \((u_{0},\omega_{0},f,g)\rightarrow (u,\omega ,p)\) is smooth. The proof essentially derives from the inverse theorem in an appropriate functional setting. The third main result proves that if \(\zeta_{0}:\mathbb{T}^{3}\rightarrow \mathbb{R}^{3}\) is a sufficiently regular conservative generator for a potential microflow \(\zeta \) as in the first result, in a neighborhood of \((0,\zeta_{0},0,0)\) in the space of data/forcing/microtorquing, the solutions to the problem are Lipschitz stable with respect to an appropriate norm. Moreover, in the case where quantitative decays of the forcing and microtorquing are given, the solutions satisfy the corresponding quantitative decay estimates.