Semiclassical states for nonlinear Dirac equations with singular potentials (Q2048876)

This article concerns the Dirac equation \[ -i \varepsilon \alpha \cdot \nabla u + a \beta u + V(x) u = K(x) |u|^{p-2} u, \quad \text{for}\ u \in H^1(\mathbb{R}^3, \mathbb{C}^4), \] where \(p\in (2,3)\), \(a>0\) is a constant, \(\alpha = (\alpha_1, \alpha_2, \alpha_3)\), and \(\alpha_1\), \(\alpha_2\), \(\alpha_3\) and \(\beta\) are \(4 \times 4\) Pauli-Dirac matrices. The novelty of this article is that the authors consider the case of singular potentials; indeed, \(V(x)\) is allowed to attain the values \(\pm a\), even at infinity, and \(K(x)\) is allowed to approach \(\pm\infty\) at points of accumulation, or even as \(|x|\rightarrow\infty\). The precise assumptions on the potentials \(V\) and \(K\), are, namely, \(V\in C(\mathbb{R}^3,\mathbb{R})\) obeys \(\sup_{\mathbb{R}^3}|V(x)|\le a\), and for \(\gamma\in(0,2),\) \[ \liminf_{|x|\rightarrow\infty} |x|^\gamma\left( a-|V(x)| \right)>0, \] and \(K\in C(\mathbb{R}^3\setminus\mathcal{K},\mathbb{R}) \cap L^{q_0}_{\mathrm{loc}}(\mathbb{R}^3,\mathbb{R})\), \(q_0>\frac{3}{3-p}\), where \(\mathcal{K}\subset\mathbb{R}^3\) is a bounded null set, and for some \(\tau\in(0,+\infty),\) there holds \[ \limsup_{|x|\rightarrow\infty}|x|^{-\tau}|K(x)| < \infty. \] A smallness condition is also made on \(\varepsilon>0\). With this, the authors provide a construction of localized bound state solutions which concentrate around the singular set of \(K\). It should be mentioned that the literature contains one comparable result (again, this only applies to the semiclassical solutions/the subcritical case), [\textit{Y. Ding} and \textit{B. Ruf}, SIAM J. Math. Anal. 44, No. 6, 3755--3785 (2012; Zbl 1259.35171)] where \(V\) is assumed to satisfy \(-a\le V(x)\le 0\) with \(\liminf_{|x|\rightarrow\infty} V(x)>-a\).