Regularizing Effect for a Nonlocal Maxwell–Schrödinger System
L. H. de Miranda, A. P. de Castro Santana · 2025
Read Regularizing Effect for a Nonlocal Maxwell–Schrödinger System by L. H. de Miranda, A. P. de Castro Santana (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-12294.txt.
Introduction
Luís Henrique de Miranda and Ayana Pinheiro de Castro Santana wrote a 2025 mathematics paper on a pair of linked equations from physics. The system joins a Schrödinger equation, which governs quantum waves, with Maxwell-type equations for electric and magnetic fields. In their version, each equation's strength depends on the whole solution at once, not just on local values. That nonlocal feature makes the system harder to solve.
The authors prove that solutions exist even when the input data are rough. They also prove a smoothing effect: the solutions turn out nicer than the data that produced them. In technical terms, the solutions gain regularity in the standard function spaces used for such problems. The result holds even though the reaction terms in the equations need not behave in a tidy, monotone way. The paper adds to the theory of coupled quantum and electromagnetic models.
Catalog record
| CREATOR | L. H. de Miranda, A. P. de Castro Santana |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 66 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-12294
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Regularizing Effect for a Nonlocal Maxwell–Schrödinger System L. H. de Miranda, A. P. de Castro Santana Abstract In this paper we prove existence and regularity of weak solutions for the following system { − div ( ( ∥ ∇ u ∥ p L p + ∥ ∇ v ∥ p L p ) | ∇ u | p − 2 ∇ u ) + g ( x , u , v ) = f in Ω ; − div ( ( ∥ ∇ u ∥ p L p + ∥ ∇ v ∥ p L p ) | ∇ v | p − 2 ∇ v ) = h ( x , u , v ) in Ω ; u = v = 0 on ∂ Ω . \displaystyle\begin{cases}&-\mbox{div}\Bigg(\bigg(\|\nabla u\|^{p}_{L^{p}}+\|\nabla v\|^{p}_{L^{p}}\bigg)|\nabla u|^{p-2}\nabla u\Bigg)+g(x,u,v)=f\ \ \ \mbox{in}\ \Omega;\\ &-\mbox{div}\Bigg(\bigg(\|\nabla u\|^{p}_{L^{p}}+\|\nabla v\|^{p}_{L^{p}}\bigg)|\nabla v|^{p-2}\nabla v\Bigg)=h(x,u,v)\ \ \ \ \mbox{in}\ \Omega;\\ &u=v=0\ \mbox{on}\ \partial\Omega.\end{cases} where Ω \Omega is an open bounded subset of ℝ N \mathbb{R}^{N} , N > 2 N>2 , f ∈ L m ( Ω ) f\in L^{m}(\Omega) , where m > 1 m>1 and g g , h h are two Carathéodory functions, which may be non monotone. We prove that under appropriate conditions on g g and h h , there is gain of Sobolev and Lebesgue regularity for the solutions of this system. Mathematics Subject Classification 2020: Primary: 35B65, 35D30; Secondary: 35B45,35D99. 1 Introduction In this paper, we investigate the existence and regularity of positive solutions for the following class of Maxwell-Schrödinger systems of Kirchhoff type (P) \displaystyle\begin{cases}&-\mbox{div}\Bigg(\bigg(\|\nabla u\|^{p}_{L^{p}}+\|\nabla v\|^{p}_{L^{p}}\bigg)|\nabla u|^{p-2}\nabla u\Bigg)+g(x,u,v)=f\ \ \ \mbox{in}\ \Omega;\\ &-\mbox{div}\Bigg(\bigg(\|\nabla u\|^{p}_{L^{p}}+\|\nabla v\|^{p}_{L^{p}}\bigg)|\nabla v|^{p-2}\nabla v\Bigg)=h(x,u,v)\ \ \ \ \mbox{in}\ \Omega;\\ &u=v=0\ \mbox{on}\ \partial\Omega,\end{cases} where, from now on, \Omega\subset\mathbb{R}^{N} , is an open bounded subset, N>2 , 1<p<N and f\in L^{m}(\Omega) for m>1 . Moreover, g and h:\Omega\times\mathbb{R}\times\mathbb{R}\to\mathbb{R} are both Carathéodory functions, which may be nonmonotone, responsible for the coupling of our system, see the assumptions below. In recent years, there have been several results investigating the gain of regularity of solutions of Maxwell–Schrödinger systems. As it turned out, even in rough regimes where the datum is not regular, the coupling in these systems allows the existence of solutions which somehow are more regular than expected, the so–called Regularizing Effect. By now, it is known that this is caused by the very interesting nature of the coupling between the equations of systems of Maxwell–Schrödinger type. Without the intention of being complete, the interested reader is invited to see [ 1 , 3 , 6 , 8 , 9 , 10 , 13 ] plus the references therein, and many others. The starting point for this analysis is due to L. Boccardo in [ 3 ] , where this phenomenon was first investigated. Later, in a series of works in collaboration with L. Orsina, other aspects of regularity gains were explored, for instance see [ 6 , 7 ] , or more recently [ 9 ] , and the references therein. Summarizing, the authors prove existence/ regularizing properties of/for solutions for a class of local Maxwell–Schrödinger systems. In order to illustrate the ideas, we quote that in [ 6 ] , they address the following problem (1.1) \displaystyle\begin{cases}&-\mbox{div}(M(x)\nabla u)+A\varphi|u|^{r-2}u=f,\ \ \ \ \ \mbox{in}\ \ \ \Omega;\\ &-\mbox{div}(M(x)\nabla\varphi)=|u|^{r},\ \ \ \ \ \ \ \ \ \ \ \mbox{in}\ \ \ \ \Omega;\\ &u=\varphi=0\ \ \ \mbox{on}\ \ \ \partial\Omega,\end{cases} where, from now on, M(x) is a uniformly elliptic symmetric bounded matrix, i.e., M(x)\xi\cdot\xi\geqslant\alpha|\xi|^{2}\mbox{ and }|M(x)|\leqslant\beta\ \ \mbox{a.e in}\ \Omega\ \mbox{and}\ \forall\ \xi\in\mathbb{R}^{N} , for 0<\alpha\leqslant\beta , A is a positive constant, and f\in L^{m}(\Omega) . In short, among other results, the authors prove that even when m<(2^{*})^{\prime} , if r>2^{*} and r^{\prime}\leqslant m , there exist u,\varphi\in W^{1,2}_{0}(\Omega) , weak solution of ( 1.1 ). This is interesting since f lies below the dual threshold (2^{*})^{\prime} , |u|^{r} is a priori low summable, and also because the weight A\varphi is not assumed to be regular or strictly positive. Moreover, when 1<r<2^{*} , if \max\{1,\frac{Nr}{N+2r}\}<m<(2^{*})^{\prime} there exists (u,\varphi)\in W^{1,m^{*}}_{0}(\Omega)\times W^{1,q}_{0}(\Omega) , solution of ( 1.1 ) for q=\begin{cases}2\mbox{ if }\frac{2Nr}{N+2+4r}\leqslant m<(2^{*})^{\prime}\\ \frac{Nm}{Nr-2mr-m}\mbox{ if }\max\{1,\frac{Nr}{N+2r}\}<m<\frac{2Nr}{N+2+4r}.\end{cases} Later on, in [ 13 ] , analogous results were obtained to the following version of ( 1.1 ) (1.2) \displaystyle\begin{cases}&-\mbox{div}(M(x)\nabla u)+g(x,u,v)=f,\ \ \ \ \ \mbox{in}\ \ \ \Omega;\\ &-\mbox{div}(M(x)\nabla v)=h(x,u,v),\ \ \ \ \ \ \ \ \ \ \ \mbox{in}\ \ \ \ \Omega;\\ &u=v=0\ \ \ \mbox{on}\ \ \ \partial\Omega,\end{cases} supposing that f\geqslant 0 a.e. in \Omega , while including two Carathéodory nonlinearities, g(x,s,t) and h(x,s,t) , satisfying: (a) there exist c_{1} , c_{2}>0 such that ( \mbox{H}^{\prime}_{1} ) \displaystyle c_{1}|s|^{r-1}|t|^{\theta+1}\leqslant|g(x,s,t)|\leqslant c_{2}|s|^{r-1}|t|^{\theta+1}; (b) g(x,s,t) is monotone in s , i.e., ( \mbox{H}_{2}^{\prime} ) \displaystyle(g(x,s_{1},t)-g(x,s_{2},t))(s_{1}-s_{2})\geqslant 0\ \ \forall\ s_{1},s_{2}\in\mathbb{R},t\geqslant 0\ \ \mbox{a.e.}\ x\ \mbox{in}\ \Omega; (c) there exist d_{1} , d_{2}>0 such that ( \mbox{H}^{\prime}_{3} ) \displaystyle d_{1}|s|^{r}|t|^{\theta}\leqslant|h(x,s,t)|\leqslant d_{2}|s|^{r}|t|^{\theta}; (d) h(.,.,.) is non-negative ( \mbox{H}^{\prime}_{4} ) \displaystyle h(x,s,t)\geqslant 0\ \ \forall\ s,t\in\mathbb{R},\ \ \mbox{a.e.}\ x\ \mbox{in}\ \Omega. In the particular case where m\geqslant(r+\theta+1)^{\prime} , 0<\theta<\min\big(\frac{4}{N-2},1\big) , 2^{*}<r+\theta+1 and (r+\theta+1)^{\prime}\leqslant m<(2^{*})^{\prime} , recalling that f\in L^{m}(\Omega) , the authors prove the existence of (u,v)\in W^{1,2}_{0}(\Omega)\times W^{1,2}_{0}(\Omega) , a regularized weak solution of ( 1.2 ). For further developments on the local case, see [ 1 ] , [ 7 ] , and [ 10 ] . In [ 1 ] , in cooperation with D. Arcoya, by assuming a interesting growth condition on f(x) , it is addressed a version of ( 1.1 ) with the constant A>0 replaced by a A(x)\in L^{1}(\Omega) , A(x)\geqslant 0 , and also with a modification of the diffusion term on the second equation. In [ 7 ] , the authors include sublinear and singular structures in ( 1.1 ), i.e., the nonlinearities become A\varphi^{\theta}u^{r-1} and \frac{u^{r}}{\varphi^{1-\theta}} where \theta\in(0,1). Moreover, for an investigation of a quasilinear version of ( 1.1 ), see [ 10 ] . Meanwhile, a nonlocal version of ( P ) was introduced by L. Boccardo and L. Orsina in [ 8 ] , where the authors addressed (1.3) \displaystyle\begin{cases}&-\mbox{div}\Big(\big(a(x)+\|\nabla u\|^{2}_{L^{2}}\big)\nabla u\Big)+\varphi|u|^{r-2}u=f\ \ \ \ \ \mbox{in}\ \ \ \Omega;\\ &-\mbox{div}(M(x)\nabla\varphi)=|u|^{r}\ \ \ \ \ \ \ \mbox{in}\ \ \ \ \ \Omega;\\ &u=v=0\ \ \ \ \ \ \mbox{on}\ \ \ \ \ \partial\Omega,\end{cases} a Kirchhoff–Schrödinger–Maxwell (KMS) system, where a:\Omega\to\mathbb{R} is a measurable function such that 0<\alpha\leqslant a(x)\leqslant\beta.\ \ x a.e in \Omega , see ( 1.1 ) above. The authors conduct a through analysis of existence, nonexistence and regularity of solutions of ( 1.3 ) for 1<m<\infty and r>1 . In particular, regarding regularizing zones, among other results, the authors proved that, if (r+1)^{\prime}\leqslant m<(2^{*})^{\prime} and r\geqslant 2^{*}-1 , or if (2^{*})^{\prime}\leqslant m and 1<r<2^{*}-1 , then there exists (u,\varphi)\in W^{1,2}_{0}(\Omega)\times W^{1,2}_{0}(\Omega) , solution of ( 1.3 ), where u\in L^{q}(\Omega) for q=\max\{{m^{**},\frac{m(2r+1)}{m+1}}\} . In the case m<(2^{*})^{\prime} , clearly q=\frac{m(2r+1)}{m+1} ensuring a double regularizing effect: there exist energy solutions despite that f\notin W^{-1,2}(\Omega) and the u has improved integrability since q>m^{**} , see Definition 1.2 . Further, let us add that as pointed out in [ 8 ] , outside the energy zone, e.g., 1<m<\min\{(r+1)^{\prime},(2^{*})^{\prime}\} , there might be interesting convergence phenomenon. Indeed,