Abstract
In this paper, we study the following nonlinear Schrödinger–Bopp–Podolsky system
where \(a>0\), \(q>0\) and \(V\in {\mathcal {C}}({\mathbb {R}}^3,{\mathbb {R}})\). By means of the variational methods, we prove the existence of infinitely many nontrivial solutions, the existence of a ground state solution for \(f(u)=|u|^{p-2}u+h(u)\) with \(p\in [4,6)\) and the existence of at least one positive solution for \(f(u)=P(x)u^5+\mu |u|^{p-2}u\) with \(p\in (2,6)\) under some certain assumptions.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
In the recent paper [12], the following Schrödinger–Bopp–Podolsky system has been studied for the first time
where \(a, \omega >0\), \(q\ne 0\) and \(p\in (2,6)\). The system appears when one couples the Schrödinger field \(\Psi =\Psi (t,x)\) with the Bopp–Podolsky Lagrangian of the electromagnetic field and considers standing wave \(\Psi (t,x)=\mathrm{e}^{iwt}u(x)\) in the purely electrostatic case. The Bopp–Podolsky theory, which is a second-order gauge theory of the electromagnetic field, was developed by Bopp [3] and then independently by Podolsky [20]. According to Mie theory [19] and its generalizations in [4,5,6,7], the Bopp–Podolsky theory was introduced to solve the alleged infinity problem in classical Maxwell theory.
d’Avenia and Siciliano [12] used variational methods to prove the existence results of problem (1.1). Indeed, the solutions can be found as critical points of a smooth energy functional. When \(p \in (2,6)\) and |q| small enough or \(p \in (3,6)\) and \(q \ne 0\), they proved that the energy functional has the mountain pass geometry; hence, the above system has a nontrivial solution. In addition, they proved the above system does not admit any nontrivial solution for \(p\ge 6\) by using Pohozaev-type identity, and in the radial case, as \(a\rightarrow 0\), the solutions they got tend to solutions of the classical Schrödinger–Poisson system. They also showed that, if \(\rho \) is the distribution density of the given charge, then the electrostatic potential \(\Phi \) satisfies the following equation
If \(\rho =4\pi \delta _{x_0}\) with \(x_0\in {\mathbb {R}}^3\), then \({\mathcal {G}}(x-x_0)=\frac{1}{|x-x_0|}\) is the fundamental solution of (1.2). And the electrostatic energy is
Hence, Eq. (1.2) is replaced by
in the Bopp–Podolsky theory. Moreover, we know that \({\mathcal {K}}(x-x_0)\) is the fundamental solution of the equation
where
And its energy is
more details can be found in [12].
Remark 1.1
The operator \(-\Delta +\Delta ^2\) appears also in other different physical and mathematical problems (see [2, 14]).
After that, Gaetano and Kaye [15] supplemented and improved some results in [12]. It has been showed that (1.1) has no solution for large values of q’s and has two radial solutions for small q’s.
Chen and Tang [10] extended the subcritical case to more general cases and dealt with the following system
where \(a>0\), \(\mu >0\), \(V\in {\mathcal {C}}({\mathbb {R}}^3,[0,\infty ))\) with \(V_\infty =\lim _{|y|\rightarrow \infty }V(y)\ge \text {sup}_{x\in {\mathbb {R}}^3}V(x)>0\), and \(f\in {\mathcal {C}}({\mathbb {R}},{\mathbb {R}})\) satisfies that there exists a constant \(p\in (2,6)\) such that \(\int ^t_0f(s){\mathrm{d}}s\ge t^p\) for all \(t\ge 0\). They obtained the existence of ground state solutions for \(p\in (4,6)\) and \(\mu >0\) or \(p\in (2,4]\) and \(\mu >\mu _*\), where \(\mu _*\) is a positive constant.
Motivated by the cited papers [10, 12], our aim here is to study the existence of ground state solutions, positive solutions and infinitely many nontrivial solutions for the following Schrödinger–Bopp–Podolsky system
where \(a>0\), \(q>0\) and V satisfies the following conditions:
- \((V_1)\):
-
\(V\in {\mathcal {C}}({\mathbb {R}}^3,{\mathbb {R}})\) and \(\inf _{x\in {\mathbb {R}}^3}V(x)>0\);
- \((V_2)\):
-
there exists a constant \(d_0>0\) such that
$$\begin{aligned} \lim \limits _{|y|\rightarrow +\infty }\text {meas}\left\{ x\in {\mathbb {R}}^3:|x-y|\le d_0,V(x)\le M\right\} =0,\quad \forall M>0. \end{aligned}$$
We work in the Hilbert space
endowed with the norm
Theorem 1.2
[23] Assume that \(\left( V_{1}\right) \)–\(\left( V_{2}\right) \) are satisfied. Then, E is continuously embedded into \(L^s({\mathbb {R}}^3)\) for \(s\in [2,6]\). Moreover, the embedding \(E\hookrightarrow L^s({\mathbb {R}}^3)\) is compact for \(s\in [2,6)\).
We first consider the following system:
Assume that f is a continuous function and satisfies the following conditions:
- \((f_{1})\):
-
\(f(t)=-f(-t)\);
- \((f_{2})\):
-
there exist \(1<\iota <5\) such that \(\lim _{|t|\rightarrow 0}\frac{f(t)}{|t|}=\lim _{|t|\rightarrow +\infty }\frac{f(t)}{|t|^\iota }=0\);
- \((f_{3})\):
-
\(\lim _{|t|\rightarrow +\infty }\frac{F(t)}{|t|^4}=\infty \), and there exists \(\mu \ge 4, \kappa >0\) such that
$$\begin{aligned} \mu F(t) \le t f(t)+\kappa t^{2}, \end{aligned}$$where \(F(t)=\int _{0}^{t} f(r) \,{\mathrm {d}} r\).
Our first result is as follows.
Theorem 1.3
Assume that \(\left( V_{1}\right) \)–\(\left( V_{2}\right) \) and \(\left( f_{1}\right) \)–\(\left( f_{3}\right) \) are satisfied. Then, the problem (1.4) possesses infinitely many nontrivial solutions.
Remark 1.4
To our knowledge, there are few results about existence of multiply solutions to Schrödinger–Bopp–Podolsky equation.
Remark 1.5
In order to assure the boundedness of the Palais–Smale sequences of the energy functional, the following condition is usually supposed.
(AR) There exists \(\mu >4\) such that
Obviously, \((f_3)\) is weaker than Ambrosetti–Rabinowitz condition.
We next consider the case of \(f(u)=|u|^{p-2}u+h(u)\), namely the following system:
where \(h\in {\mathcal {C}}({\mathbb {R}},{\mathbb {R}})\) satisfies the following conditions:
- \(\left( h_{1}\right) \):
-
for some \(\varrho \in (1,5)\),
$$\begin{aligned} \lim \limits _{|u| \rightarrow +\infty } \frac{h(u)}{|u|^{\varrho }}=\lim \limits _{|u| \rightarrow 0} \frac{h(u)}{|u|}=0; \end{aligned}$$ - \(\left( h_{2}\right) \):
-
for any \(p\in [4,6)\), \(H(u)+\frac{1}{p}|u|^p\ge 0\), where \(H(u)=\int _{0}^{u} h(r) \,{\mathrm {d}} r\),
- \(\left( h_{3}\right) \):
-
there exists \(\theta _0\in (0,1)\) such that for any \(t>0\) and \(u\ne 0\),
$$\begin{aligned} \left[ \frac{h(u)}{u^3}-\frac{h(tu)}{(tu)^3}\right] \text {sign}(1-t)+\theta _0\frac{V(x)|1-t^2|}{(tu)^2}\ge 0. \end{aligned}$$
The following is the second result of our paper.
Theorem 1.6
Assume that \(\left( V_{1}\right) \)–\(\left( V_{2}\right) \) and \(\left( h_{1}\right) \)–\(\left( h_{3}\right) \) are satisfied. If \(p\in [4,6)\), then the problem (1.5) has a ground state solution.
Remark 1.7
The condition \((h_3)\) comes from [11], and it is weaker than Ambrosetti–Rabinowitz condition and also weaker than the following condition:
Finally, we consider the case of \(f(u)=P(x)u^5+\mu |u|^{p-2}u\), namely the following system:
We assume that P(x) is a continuous function and satisfies the following conditions:
- \(\left( P_{1}\right) \):
-
\(P(x)> 0\) and \(P(x)\in L^{\infty }({\mathbb {R}}^3)\);
- \(\left( P_{2}\right) \):
-
\(|P(x)-P(x_0)|=O(|x-x_0|^2)\) as \(x\rightarrow x_0\), where \(P(x_0)=\sup _{x\in {\mathbb {R}}^3}P(x)\).
The difficulty in obtaining solutions of the problem (1.6) lies in two aspects. On the one hand, since the problem involves critical exponent, the difficulty lies in the lack of compactness. On the other hand, when \(p\in (2,4)\), the boundedness of the Palais–Smale sequences of the energy functional is hard to get. Inspired by [1, 9, 12, 13, 16, 21], we construct a truncated function and use the mountain pass theorem with mountain pass lever value under \(\frac{1}{3}S^{\frac{3}{2}}\Vert P(x)\Vert ^{\frac{1}{2}}_\infty \).
Now we are in a position to state the following results of our paper on the existence of positive solutions.
Theorem 1.8
Assume that \(\left( V_{1}\right) \)–\(\left( V_{2}\right) \) and \(\left( P_{1}\right) \)–\(\left( P_{2}\right) \) are satisfied. If \(p\in (4,6)\) with \(\mu >0\) or \(p=4\) with \(\mu \) sufficiently large, then the problem (1.6) has at least one positive solution for any \(q>0\).
Theorem 1.9
Assume that \(\left( V_{1}\right) \)–\(\left( V_{2}\right) \) and \(\left( P_{1}\right) \)–\(\left( P_{2}\right) \) are satisfied. If \(p\in (2,4)\) with \(\mu \) sufficiently large, then there exists \(q_*>0\) such that, for all \(q\in (0,q_*)\), the problem (1.6) has at least one positive solution.
Notations Throughout the paper, we denote by \(\text {meas}(\cdot )\) the Lebesgue measure in \({\mathbb {R}}^3\). \(E^*\) is the dual space of E. \(\left\| \cdot \right\| _p\) denotes the usual norm of Lebesgue space \(L^p({\mathbb {R}}^3)\). S is the best Sobolev constant for the embedding \(D^{1,2}({\mathbb {R}}^3)\hookrightarrow L^6({\mathbb {R}}^3)\). The symbol C denotes a positive constant and may vary from line to line.
2 Preliminary
In this section, we want to introduce the variational setting, the functional setting and some main results.
2.1 The Variational Setting
We first consider the nonlinear Schrödinger Lagrangian density
where \(\psi :{\mathbb {R}}\times {\mathbb {R}}^3\rightarrow {\mathbb {C}}\) and \({\hbar }\),\(m>0\). If the gauge potential of the electromagnetic field \(({\mathbf {E}},{\mathbf {H}})\) is \((\phi ,{\mathbf {A}})\), where \(\phi :{\mathbb {R}}^3\rightarrow {\mathbb {R}}\) and \({\mathbf {A}}:{\mathbb {R}}^3\rightarrow {\mathbb {R}}^3\), then the following equations hold
To study the interaction between \(\psi \) and its electromagnetic field \(({\mathbf {E}},{\mathbf {H}})\), we can replace the derivatives \(\partial _t\) and \(\nabla \) in (2.1) with the covariant ones
Thus, we have
According to [3], the Bopp–Podolsky Lagrangian density is
Now, we add the Lagrangian density of the electromagnetic field to \({\mathcal {L}}_{\mathrm{CSc}}\) so that we can get the total Lagrangian density
Therefore, the total action is
More details can be found in [12].
2.2 The Operator \(-\Delta +a^2\Delta ^2\)
Let \({\mathcal {D}}\) be the completion of \(C_c^\infty ({\mathbb {R}}^3)\) with the inner product
Obviously, \({\mathcal {D}}\) is continuously embedded into \(D^{1,2}({\mathbb {R}}^3)\) and therefore in \(L^6({\mathbb {R}}^3)\).
The following lemmas were obtained in [12].
Lemma 2.1
[12] The Hilbert space \({\mathcal {D}}\) is continuously embedded in \(L^\infty ({\mathbb {R}}^3)\).
Lemma 2.2
[12] The space \(C_c^\infty ({\mathbb {R}}^3)\) is dense in \({\mathcal {A}}\), where
normed by \(\sqrt{\langle \phi ,\phi \rangle _{\mathcal {D}}}\) and, consequently \({\mathcal {D}}={\mathcal {A}}\).
Lemma 2.3
[12] For all \(y\in {\mathbb {R}}^3\), \({\mathcal {K}}(\cdot -y)=\frac{1-\mathrm{e}^{-|\cdot -y|/a}}{|\cdot -y|}\) is the fundamental solution of
Furthermore,
-
(i)
if \(f\in L^1_{\mathrm{loc}}({\mathbb {R}}^3)\) and, for almost everywhere \(x\in {\mathbb {R}}^3\), the map \(y\in {\mathbb {R}}^3\mapsto \frac{f(y)}{|x-y|}\) is summable, then \({\mathcal {K}}*f\in L^1_{\mathrm{loc}}({\mathbb {R}}^3)\);
-
(ii)
if \(f\in L^s({\mathbb {R}}^3)\) with \(1\le s < \frac{3}{2}\), then \({\mathcal {K}}*f\in L^r({\mathbb {R}}^3)\) for \(r\in \left( \frac{3s}{3-2s},+\infty \right] .\)
In both cases, \({\mathcal {K}}*f\) solves
Moreover, almost everywhere in \({\mathbb {R}}^3\), we have
Then, for any fixed \(u\in H^1({\mathbb {R}}^3)\), \(\phi _u:={\mathcal {K}}*u^2\) is the unique solution in \({\mathcal {D}}\) of the second equation in (1.3), and the following lemma holds.
Lemma 2.4
[12] For every \(u\in H^1({\mathbb {R}}^3)\), we have:
-
(i)
for every \(y\in {\mathbb {R}}^3\), \(\phi _{u(\cdot +y)}=\phi _u(\cdot +y)\);
-
(ii)
\(\phi _u\ge 0\);
-
(iii)
for every \(s\in (3,+\infty ]\), \(\phi _u\in L^s({\mathbb {R}}^3)\bigcap C_0({\mathbb {R}}^3)\);
-
(iv)
for every \(s\in \left( \frac{3}{2},+\infty \right] \), \(\nabla \phi _u=\nabla {\mathcal {K}}*u^2\in L^s({\mathbb {R}}^3)\bigcap C_0({\mathbb {R}}^3)\);
-
(v)
\(\phi _u\in {\mathcal {D}}\);
-
(vi)
\(\Vert \phi _u\Vert _6\le C\Vert u\Vert ^2\);
-
(vii)
\(\phi _u\) is the unique minimizer of the functional
$$\begin{aligned} E(\phi )=\frac{1}{2}\Vert \nabla \phi \Vert ^2_2+\frac{a^2}{2}\Vert \Delta \phi \Vert _2^2-\int _{{\mathbb {R}}^3}\phi u^2, \quad \phi \in {\mathcal {D}}; \end{aligned}$$ -
(viii)
if \(v_n\rightharpoonup v\) in \(H^1({\mathbb {R}}^3)\), then \(\phi _{v_n}\rightharpoonup \phi _v\) in \({\mathcal {D}}\).
2.3 The Functional Setting
It is easy to see that the weak solutions of (1.3) are the critical points of the \(C^1\) energy functional
on \(E\times {\mathcal {D}}\). If \((u,\phi )\in E\times {\mathcal {D}}\) is a critical point of R, then we have
and
Distinctly, the functional R is strongly indefiniteness, and hence, we adopt a reduction procedure which used in [12]. Noting that \(\partial _\phi R\in {\mathcal {C}}^1({\mathbb {R}}^3)\), by implicit function theorem, we have
where \(G_\Phi \) is the graph of the map \(\Phi :u\in E\mapsto \phi _u\in {\mathcal {D}}\). Hence, we have the reduced form
and it is of class \({\mathcal {C}}^1\) on E. For all \(u,v\in E\),
In fact, to find solutions of Eq. (1.3), we only need to find critical points of I.
3 Existence of Multiple Solutions
In this section, assume that \(\left( V_{1}\right) \)–\(\left( V_{2}\right) \) and \(\left( f_{1}\right) \)–\(\left( f_{3}\right) \) are satisfied. We prove that the problem (1.4) possesses infinitely many nontrivial solutions. And we give the energy functional I by
Lemma 3.1
[22] Let X be an infinite-dimensional Banach space, and there exists a finite-dimensional space W such that \(X=W \oplus V\). \(I \in {\mathcal {C}}^{1}(X,{\mathbb {R}})\) satisfies the (PS) condition, and
-
(i)
\(I(u)=I(-u)\) for all \(u \in X\), \(I(0)=0\);
-
(ii)
there exist \(\rho >0\), \(\alpha >0\) such that \(\left. I\right| _{\partial B_{\rho } \cap V} \ge \alpha \);
-
(iii)
for any finite-dimensional subspace \(Y \subset X\), there exists \(R=R(Y)>0\) such that \(I(u) \le 0\) on \(Y \backslash B_{R}\).
Then, I possesses an unbounded sequence of critical values.
Lemma 3.2
Under the assumptions of Theorem 1.3, if \(\left\{ u_{n}\right\} \subset E\) satisfies the following conditions:
-
(i)
\(\left\{ I\left( u_{n}\right) \right\} \) is bounded above by a positive constant;
-
(ii)
\(I^{\prime }\left( u_{n}\right) [u_{n}] \rightarrow 0\), then \(\left\{ u_{n}\right\} \) is bounded in E.
Proof
If \(\left\{ u_{n}\right\} \) is unbounded in E, we can find a subsequence still denoted by \(\left\{ u_{n}\right\} \) such that \(\Vert u_{n}\Vert \rightarrow +\infty \). Let \(v_{n}=\frac{u_{n}}{\left\| u_{n}\right\| }\), we have \(\left\| v_{n}\right\| =1\). Thus, we may assume that \(v_{n}\rightharpoonup v\) in E. It follows from \(\left( f_{3}\right) \) and (i) that there exists a constant \(c>0\) such that
as \(n \rightarrow \infty \), which implies \(1 \le \frac{2 \kappa }{\mu -2} \limsup _{n \rightarrow \infty }\left\| v_{n}\right\| _{2}^{2}\). Therefore, \(v \not \equiv 0\).
By \(\left( f_{3}\right) \) and Fatou’s lemma, one has
as \(n\rightarrow +\infty \). This is a contradiction, and hence, \(\left\{ u_{n}\right\} \) is bounded in E. \(\square \)
Lemma 3.3
Under the assumptions of Theorem 1.3, then the functional I satisfies the (PS) condition.
Proof
To prove that I satisfies the (PS) condition, we only need to prove \(\{u_n\}\subset E\) which obtained by Lemma 3.2, has a convergent subsequence. As \(\{u_n\}\) is bounded in E, there exists a subsequence still denoted by \(\{u_n\}\) and \(u_0\in E\) such that \(u_n\rightharpoonup u_0\) in E and \(u_n\rightarrow u_0\) in \(L^q({\mathbb {R}}^3)\) for \(2< q<6\). It follows from \(u_n\rightharpoonup u_0\) in E and \(I'(u_0)\in E^*\) that \(\left\langle I^{\prime }(u_0), u_{n}-u_0\right\rangle \rightarrow 0\). And as \(I'(u_n)\rightarrow 0\) in \(E^*\), it is easy to obtain
Therefore,
By \((f_2)\), for any \(\epsilon >0\), there exists \(C_\epsilon >0\) such that
Hence, as \(n\rightarrow +\infty \), we have
as \(\epsilon \rightarrow 0\) and \(\iota \in (1,5)\).
And then,
This completes the proof.\(\square \)
Proof of Theorem 1.3
We have verified that I satisfies the (PS) condition. It follows from \((f_1)\) that I is an even functional. As E is a separable space, E has orthonormal basis \(\{e_i\}\). For all \(k,j\in {\mathbb {Z}}\), we define \(E_{j}:={\mathbb {R}} e_{j}\), \(W_{k}:=\oplus _{j=1}^{k} E_{j}\), \(V_{k}:=\overline{\oplus _{j=k+1}^{\infty }E_{j}}\). Let \(W=W_{k}\), \(V=V_{k}\), clearly \(E=W \oplus V\) and \({\text {dim}} W<\infty \).
Next, we verify that I satisfies (ii) in Lemma 3.1. By \((f_2)\), for any \(\epsilon >0\), there exists \(C_\epsilon >0\) such that
Therefore, there exists \(\rho >0\) small enough, \(\alpha >0\) such that \(I(u) \ge \alpha >0\) as \(\Vert u\Vert =\rho \).
Now, we verify that I satisfies (iii) in Lemma 3.1. If \(I(u)>0\) for any finite-dimensional subspace \(Y\subset E\), then we can find a sequence \(\{u_n\}\) such that \(\Vert u_n\Vert \rightarrow +\infty \) and \(I(u_n)>0\). Let \(v_{n}=\frac{u_{n}}{\left\| u_{n}\right\| }\), from the proof in Lemma 3.2, we have
as \(n\rightarrow +\infty \), which is a contradiction. This shows that there exists \(R=R(Y)>0\) such that \(I(u) \le 0\) on \(Y \backslash B_{R}\). The proof is completed.\(\square \)
4 Existence of Ground State Solutions for \(p\in [4,6)\)
In this section, to solve the problem (1.5), we apply Jeanjean’s trick [17] and give a family of energy functionals
Lemma 4.1
[24] Assume that \(\left\{ u_{n}\right\} \) is a bounded sequence in \(H^1({\mathbb {R}}^3)\). If
then \(u_{n} \rightarrow 0\) in \(L^{q}\left( {\mathbb {R}}^{3}\right) \) for all \(q \in (2, 6)\).
Lemma 4.2
[17] Let \((X,\Vert \cdot \Vert )\) be a Banach space and \(T\subset {{\mathbb {R}}}^+\) be an interval. Consider a family of \({\mathcal {C}}^{1}\) functionals on X of the form
with \(B(u) \ge 0\) and either \(A(u) \rightarrow +\infty \) or \(B(u) \rightarrow +\infty \) as \(\Vert u\Vert \rightarrow +\infty \). If there are two points \(v_{1}, v_{2} \in X\) such that
where
Then, for almost every \(\lambda \in T\), there exists a bounded \((PS)_{c_{\lambda }}\) sequence in X, and the mapping \(\lambda \rightarrow c_{\lambda }\) is non-increasing and continuous from the left.
Lemma 4.3
Suppose that V, h satisfy \((V_1)\)–\((V_2)\) and \((h_1)\)–\((h_2)\). Then, for almost every \(\lambda \in [\frac{1}{2},1]\), there is a bounded sequence \(\{v_m\}\), such that \(I_\lambda (v_m)\rightarrow c_\lambda \) in E and \(I'_\lambda (v_m)\rightarrow 0\) in the dual \(E^*\) of E.
Proof
Obviously that
and
Let \(u_\tau =\tau ^2u(\tau \cdot )\), then
Since \(p\in [4,6)\), there exists \(\tau _{0}>0\) large enough such that \(I_{\lambda }\left( \tau _{0} u\right) <0\). By taking \(v=u_{\tau _o}\), we have \(I_{\lambda }(v)<0\).
Moreover, it follows from \((h_1)\) that
By simple calculations, we derive
Hence,
Then, there exists \(\rho >0\) small enough such that \(I_\lambda (u)>0\) as \(\Vert u\Vert =\rho \). Now, Lemma 4.2 leads to the conclusion. \(\square \)
Lemma 4.4
If \(\left\{ v_{m}\right\} \) is a bounded sequence in E obtained by Lemma 4.3, then
Proof
Assume that \(\lim _{m \rightarrow +\infty } \sup \limits _{y \in {\mathbb {R}}^{3}}\int \limits _{B_{1}( y)}\left| v_{m}\right| ^{2}{\mathrm{d}}x=0\). From Lemma 4.1, we have \(v_{m} \rightarrow 0\) in \(L^{q}\left( {\mathbb {R}}^{3}\right) \) for all \(q \in (2, 6)\).
Hence,
Moreover, by the Hardy–Littlewood–Sobolev inequality (see [18]), we obtain
Thus, we have
and then
which is a contradiction.\(\square \)
Lemma 4.5
If \(\left\{ v_{m}\right\} \subset E\) is the sequence obtained by Lemma 4.3, then for a.e. \(\lambda \in \left[ \frac{1}{2}, 1\right] \), there exists a sequence of points \(\left\{ y_{m}\right\} \subset {\mathbb {R}}^{3}\), set \(u_{m}(x):=v_{m}\left( x-y_{m}\right) \), we have
-
(i)
\(u_{m}\rightharpoonup u_{\lambda } \not \equiv 0\) in E;
-
(ii)
\(I_{\lambda }^{\prime }\left( u_{\lambda }\right) =0\) in \(E^{*}\);
-
(iii)
\(I_{\lambda }\left( u_{\lambda }\right) \le c_{\lambda }\) in E;
-
(iv)
there exists \(M>0\) such that \(I_{\lambda }\left( u_{\lambda }\right) \ge M\).
Proof
Putting together Lemmas 4.3 and 4.4, we know that for almost every \(\lambda \in \left[ \frac{1}{2}, 1\right] \), there exists a bounded sequence \(\left\{ v_{m}\right\} \) which satisfies \(I_{\lambda }\left( v_{m}\right) \rightarrow c_{\lambda }\) in E and \(I_{\lambda }^{\prime }\left( v_{m}\right) \rightarrow 0\) in \(E^{*}\) as \(m \rightarrow +\infty \). Furthermore, by Lemma 4.4, there exists a sequence of points \(\left\{ y_{m}\right\} \subset {\mathbb {R}}^3 \) and \(\alpha >0\), such that
Let \(u_{m}(x):=v_{m}\left( x-y_{m}\right) \). By the invariance translations of \(I_{\lambda }\), as \(m \rightarrow +\infty \), we have that \(I_{\lambda }\left( u_{m}\right) \rightarrow c_{\lambda }\) in E and \(I_{\lambda }^{\prime }\left( u_{m}\right) \rightarrow 0\) in \(E^{*}\). Since \(\left\{ u_{m}\right\} \) is bounded, there exists \(u_{\lambda } \in E\) such that
In the following, we complete the proof of this lemma.
-
(i)
It follows from (4.1) that
$$\begin{aligned} C\left\| u_{\lambda }\right\| ^{2} \ge \left\| u_{\lambda }\right\| _{2}^{2} \ge \int _{B_{1}(0)} u_{\lambda }^{2} \,{\mathrm {d}} x=\lim \limits _{m \rightarrow +\infty } \int _{B_{1}(0)} u_{m}^{2} \,{\mathrm {d}} x \ge \alpha >0, \end{aligned}$$and thus \(u_{\lambda } \not \equiv 0\) in E.
-
(ii)
We only need to show that \(\left\langle I_{\lambda }^{\prime }\left( u_{\lambda }\right) , \varphi \right\rangle = 0\) for any \(\varphi \in E\). Observe that
$$\begin{aligned}&\left\langle I_{\lambda }^{\prime }\left( u_{m}\right) , \varphi \right\rangle -\left\langle I_{\lambda }^{\prime }\left( u_{\lambda }\right) , \varphi \right\rangle \\&\quad =\int _{{\mathbb {R}}^3}\left( \left( \nabla u_m-\nabla u_\lambda \right) \varphi +V(x)\left( u_m-u_\lambda \right) \varphi \right) \,{\mathrm {d}} x+\int _{{\mathbb {R}}^3}\left( \phi _{u_m}u_m-\phi _{u_\lambda }u_\lambda \right) \varphi \,{\mathrm {d}} x\\&\qquad -\,\lambda \int _{{\mathbb {R}}^3}\left( f(u_m)-f(u_\lambda )\right) \varphi \,{\mathrm {d}} x-\lambda \int _{{\mathbb {R}}^3}\left( |u_m|^{p-2}u_m-|u_\lambda |^{p-2}u_\lambda \right) \varphi \,{\mathrm {d}} x. \end{aligned}$$Moreover,
$$\begin{aligned} \left| \int _{{\mathbb {R}}^3}\left( \phi _{u_m}u_m-\phi _{u_\lambda }u_\lambda \right) \varphi \,{\mathrm {d}} x\right|\le & {} \int _{{\mathbb {R}}^3}\phi _{u_m}|u_m-u_{\lambda }||\varphi |\,{\mathrm {d}} x\\&+\,\int _{{\mathbb {R}}^3}|\phi _{u_m}-\phi _{u_\lambda }||u_\lambda \varphi |\,{\mathrm {d}} x:=J_1+J_2. \end{aligned}$$By Lemma 2.4, \(\{\phi _{u_m}\}\) is bounded in \(L^6({\mathbb {R}}^3)\) and \(\phi _{u_m}\rightharpoonup \phi _{u_\lambda }\) in \(L^6({\mathbb {R}}^3)\). Using the Hölder inequality and the strong convergence of \(u_m\) to \(u_\lambda \) in \(L^3({\mathbb {R}}^3)\), we get
$$\begin{aligned} J_1\le \Vert \phi _{u_m}\Vert _6\Vert u_m-u_\lambda \Vert _{3}\Vert \varphi \Vert _{2}\rightarrow 0. \end{aligned}$$Since \(\varphi \in E\), for any \(\epsilon >0\) there exists \(R>0\) such that
$$\begin{aligned} \left( \int _{B_R^c(0)}|\varphi |^s\,{\mathrm {d}} x\right) ^{\frac{1}{s}}<\epsilon ,\quad s\in [2,6]. \end{aligned}$$(4.3)Hence,
$$\begin{aligned} J_2= & {} \int _{B_R(0)}|\phi _{u_m}-\phi _{u_\lambda }||u_\lambda \varphi |\,{\mathrm {d}} x+\int _{B^c_R(0)}|\phi _{u_m}-\phi _{u_\lambda }||u_\lambda \varphi |\,{\mathrm {d}} x\\\le & {} \Vert \phi _{u_m}-\phi _{u_\lambda }\Vert _{L^6(B_R(0))}\Vert u_\lambda \Vert _2\Vert \varphi \Vert _3+\Vert \phi _{u_m}-\phi _{u_\lambda }\Vert _{6}\Vert u_\lambda \Vert _2\Vert \varphi \Vert _{L^3(B_R^c(0))}\\\rightarrow & {} 0, \end{aligned}$$as \(m\rightarrow +\infty \) and \(R\rightarrow +\infty \). Furthermore, notice that
$$\begin{aligned}&\left| \int _{{\mathbb {R}}^3}\left( |u_m|^{p-2}u_m-|u_\lambda |^{p-2}u_\lambda \right) \varphi \,{\mathrm {d}}x\right| \\&\quad \le \int _{B_R(0)}\left| |u_m|^{p-2}u_m-|u_\lambda |^{p-2}u_\lambda \right| \left| \varphi \right| \,{\mathrm {d}} x\\&\quad \quad +\,\int _{B^c_R(0)}\left| |u_m|^{p-2}u_m-|u_\lambda |^{p-2}u_\lambda \right| \left| \varphi \right| \,{\mathrm {d}} x\\&\quad :=J_3+J_4. \end{aligned}$$From (4.2), we infer that there exists \(Q(x)\in L^p(B_R(0))\) such that \(|u_m|\le Q(x)\). It follows from dominated convergence theorem that \(J_3\rightarrow 0\) as \(m\rightarrow 0\). On the other hand, as \(R\rightarrow \infty \), we obtain
$$\begin{aligned} J_4\le & {} \int _{B^c_R(0)}|u_m|^{p-1}\left| \varphi \right| +\int _{B^c_R(0)}|u_\lambda |^{p-1}\left| \varphi \right| \,{\mathrm {d}} x\\\le & {} \Vert u_m\Vert _p^{p-1}\Vert \varphi \Vert _{L^p(B^c_R(0))}+\Vert u_\lambda \Vert _p^{p-1}\Vert \varphi \Vert _{L^p(B^c_R(0))}\\\le & {} \epsilon C\rightarrow 0. \end{aligned}$$Moreover, as \(u_{m} \rightarrow u_{\lambda }\ \,\mathrm{a.e.\, in}\,{\mathbb {R}}^3\), by continuity,
$$\begin{aligned} h(u_m)\rightarrow h(u_\lambda )\quad \text {a.e.}\, \text {in}\quad {\mathbb {R}}^3. \end{aligned}$$It follows from \((h_1)\) that \(\{h(u_m)\}\) is bounded in \(L^{\frac{6}{5}}({\mathbb {R}}^3)\), and so \(h(u_m)\rightharpoonup h(u_\lambda )\) in \(L^{\frac{6}{5}}({\mathbb {R}}^3)\). Hence, as \(R\rightarrow \infty \), we have
$$\begin{aligned}&\left| \int _{{\mathbb {R}}^3}\left( h(u_m)-h(u_\lambda )\right) \varphi \,{\mathrm {d}} x\right| \\&\quad \le \int _{B_R(0)}\left| h(u_m)-h(u_\lambda )\right| \left| \varphi \right| \,{\mathrm {d}} x+\int _{B^c_R(0)}\left| h(u_m)-h(u_\lambda )\right| \left| \varphi \right| \,{\mathrm {d}}x\\&\quad \le \int _{B_R(0)}\left| h(u_m)-h(u_\lambda )\right| \left| \varphi \right| \,{\mathrm {d}} x+\int _{B^c_R(0)}|h(u_m)|\left| \varphi \right| \,{\mathrm {d}} x\\&\qquad +\,\int _{B^c_R(0)}|h(u_\lambda )|\left| \varphi \right| \,{\mathrm {d}} x\\&\quad \le \Vert h(u_m)-h(u_\lambda )\Vert _{L^{\frac{6}{5}}(B_R(0))}\Vert \varphi \Vert _6+\Vert u_m\Vert _{\frac{6}{5}}\Vert \varphi \Vert _{L^6(B^c_R(0))}\\&\qquad +\,\Vert u_\lambda \Vert _{\frac{6}{5}}\Vert \varphi \Vert _{L^6(B^c_R(0))}\\&\quad \le o(1)+\epsilon C\rightarrow 0. \end{aligned}$$Now, we get \(I'_\lambda (u_\lambda )=0\).
-
(iii)
It follows from \(\left( h_{3}\right) \) that
$$\begin{aligned}&\frac{1}{4}h(u)u-H(u)+\frac{\theta _0V(x)}{4}u^2\\&\quad =\int ^1_0\left[ \frac{h(u)}{u^3}-\frac{h(su)}{{su}^3}+\frac{\theta _0V(x)(1-s^2)}{(su)^2}\right] s^3u^4{\mathrm{d}}s\ge 0. \end{aligned}$$Hence, by Fatou’s lemma we get
$$\begin{aligned} c_{\lambda }= & {} \lim \limits _{m \rightarrow +\infty }\left[ I_{\lambda }\left( u_{m}\right) -\frac{1}{4}\left\langle I_{\lambda }^{\prime }\left( u_{m}\right) , u_{m}\right\rangle \right] \\\ge & {} \left( \frac{\lambda }{4}-\frac{\lambda }{p}\right) \liminf _{m \rightarrow +\infty }\Vert u_m\Vert _p^p+ \frac{1}{4}\liminf _{m \rightarrow +\infty }\left( \Vert u_m\Vert ^2-\lambda \theta _0\int _{{\mathbb {R}}^3}V(x)|u_m|^2{\mathrm{d}}x\right) \\&+\,\lambda \liminf _{m \rightarrow +\infty }\int _{{\mathbb {R}}^3}\left( \frac{1}{4}h(u_m)u_m-H(u_m)+\frac{\theta _0V(x)|u_m|^2}{4}\right) {\mathrm{d}}x\\\ge & {} \left( \frac{\lambda }{4}-\frac{\lambda }{p}\right) \Vert u_\lambda \Vert _p^p+ \frac{1}{4}\left( \Vert u_\lambda \Vert ^2-\lambda \theta _0\int _{{\mathbb {R}}^3}V(x)|u_\lambda |^2{\mathrm{d}}x\right) \\&+\,\lambda \int _{{\mathbb {R}}^3}\left( \frac{1}{4}h(u_\lambda )u_\lambda -H(u_\lambda )+\frac{\theta _0V(x)|u_\lambda |^2}{4}\right) {\mathrm{d}}x\\= & {} I_{\lambda }\left( u_{\lambda }\right) -\frac{1}{4}\left\langle I_{\lambda }^{\prime }\left( u_{\lambda }\right) , u_{\lambda }\right\rangle =I_{\lambda }\left( u_{\lambda }\right) >0. \end{aligned}$$ -
(iv)
Recall that \(C\Vert u_\lambda \Vert ^2\ge \alpha \), we infer
$$\begin{aligned} I_{\lambda }\left( u_{\lambda }\right) \ge \frac{1-\theta _0}{4}\Vert u_\lambda \Vert ^2\ge \frac{\alpha }{C}:=M. \end{aligned}$$
The proof is completed.\(\square \)
Now, according to Lemmas 4.3 and 4.5, there exists a sequence \(\left\{ \left( \lambda _{n}, u_{\lambda _{n}}\right) \right\} \subset \left[ \frac{1}{2}, 1\right] \times E\), such that
-
(i)
\(\lambda _{n} \rightarrow 1\) as \(n \rightarrow +\infty \);
-
(ii)
\(u_{\lambda _{n}} \not \equiv 0\), \(M \le I_{\lambda _{n}}\left( u_{\lambda _{n}}\right) \le c_{\lambda _{n}}\);
-
(iii)
\(I_{\lambda _{n}}^{\prime }\left( u_{\lambda _{n}}\right) =0\).
Proof of Theorem 1.6
By Lemma 4.5, we have
which implies that there exists a constant \(K>0\) such that\(\Vert u_{\lambda _{n}}\Vert \le K.\)
Using the facts that for any \(\varphi \in E\),
and \(\left\{ u_{\lambda _{n}}\right\} \) is bounded in E, we obtain that \(M \le \lim _{n \rightarrow +\infty } I\left( u_{\lambda _{n}}\right) \le c_{1}\) and \(\lim _{n \rightarrow +\infty } I^{\prime }\left( u_{\lambda _{n}}\right) =0\). Up to a subsequence, there exists a subsequence still denoted by \(\left\{ u_{\lambda _{n}}\right\} \) and \(u_{0} \in E\) such that \(u_{\lambda _{n}}\rightharpoonup u_{0}\) in E. By using the preceding method in Lemma 4.5, we can obtain the existence of a nontrivial solution \(u_{0}\) for I such that \(I^{\prime }\left( u_{0}\right) =0\) and \(M\le I\left( u_{0}\right) \le c_{1}\). Thus, \(u_{0}\) is a nontrivial solution of (1.5). Define \(m:=\inf \left\{ I(u): u \not \equiv 0, I^{\prime }(u)=0\right\} \). Let \(\left\{ u_{n}\right\} \) be a sequence such that \(I^{\prime }\left( u_{n}\right) =0\) and \(I\left( u_{n}\right) \rightarrow m\). Similar to the argument in Lemma 4.5, we can prove that there exists \({\bar{u}} \in E\) such that \(I^{\prime }({\bar{u}})=0\) and \(I({\bar{u}}) \le m\). By the definition of m, we have \(m \le I({\bar{u}})\). Hence, \(I({\bar{u}})=m\), which shows that \({\bar{u}}\) is a ground state solution of (1.5).\(\square \)
5 Existence of Positive Solutions for \(p\in (2,6)\)
In this section, assume that \(\left( V_{1}\right) \)–\(\left( V_{2}\right) \) and \(\left( P_{1}\right) \)–\(\left( P_{2}\right) \) are satisfied and \(p\in (2,6)\). We prove the existence of positive solutions for problem (1.6).
By simple calculation, we have
Lemma 5.1
[12] If \(v_n\rightharpoonup v\) in \(H^1({\mathbb {R}}^3)\), then
Lemma 5.2
Let \(\{v_n\}\) be a sequence such that \(v_n\rightharpoonup v\) in E, then
Proof
To prove \(\int _{{\mathbb {R}}^3}\phi _{v_n}v^2_n\rightarrow \int _{{\mathbb {R}}^3}\phi _{v}v^2\), we only need to prove that
We just give the proof of \(J_1\rightarrow 0\), and the proof of \(J_3\rightarrow 0\) can be proved in the same way. Lemma 2.4 and Theorem 1.2 imply that
Moreover, since \(v\in E\), for any \(\epsilon >0\) there exists \(R>0\) such that
Obviously,
as \(m\rightarrow +\infty \) and \(R\rightarrow +\infty \), and \(J_4\rightarrow 0\) can be proved in the same way. \(\square \)
5.1 The Case \(p\in [4,6)\)
Lemma 5.3
Assume that \(\left( V_{1}\right) \)–\(\left( V_{2}\right) \) and \((P_1)\)–\(\left( P_{2}\right) \) are satisfied. Then, the functional I verifies the mountain pass geometry, that is,
-
(i)
there exist \(\alpha ,\rho >0\) such that \(I(v)\ge \alpha \) for all \(\Vert v\Vert =\rho \);
-
(ii)
there exists \(e\in H^1(\mathbb (R)^3){\setminus }\{0\}\) such that \(I(e)<0\) with \(\Vert e\Vert >\rho \).
Proof
Since \(P(x)\le P(x_0)\), we have
Then, there exist \(\alpha >0\) and \(\rho >0\) small enough such that \(I(u)\ge \alpha \) for all \(\Vert u\Vert =\rho \).
Moreover, let \(u_\tau =\tau u\), then
Hence, there exists \(\tau _{0}>0\) large enough such that \(I\left( \tau _{0} u\right) <0\). By taking \(e=u_{\tau _o}\), we have \(I(e)<0\). \(\square \)
Combine Lemma 5.3 and Mountain–Pass lemma, we infer that there exists a sequence \(\{v_n\}\) in E at the level c, such that
where
Lemma 5.4
The sequence \(\{v_n\}\) in (5.1) is bounded in E for \(p\in [4,6)\).
Proof
As \(n\rightarrow +\infty \), we have
This prove Lemma 5.4.\(\square \)
Lemma 5.5
The sequence \(\{v_n\}\) in (5.1) is compact in E if
Proof
As the sequence \(\{v_n\}\) is given by (5.1), it satisfies that
Since the sequence \(\{v_n\}\) is bounded, there exists \(v \in E{\setminus }\{0\}\) such that
From \(I'(v_n)\rightarrow 0\), we can obtain
Set \(v'_n=v_n-v\). By Brezis–Lieb’s lemma in [8],
Theorem 1.2 and Lemmas 5.1, 5.2 and (5.4) imply that
Putting (5.8) in (5.6) and (5.7), as \(n\rightarrow +\infty \), we have
and
Assume that \(\int _{{\mathbb {R}}^3}P(x)|v'_n|^6{\mathrm{d}}x\rightarrow l\) as \(n\rightarrow +\infty \), then \(\int _{{\mathbb {R}}^3}|\nabla v'_n|^2+V(x)|v'_n|^2{\mathrm{d}}x\rightarrow l\).
It follows from Sobolev inequality that
If \(l>0\), we have
(5.9), (5.10) and Lemma 5.4 imply that
However, (5.5) shows that
which is a contradiction to (5.11), thus \(l=0\). This concludes that \(\{v_n\}\) is compact.
\(\square \)
Proof of Theorem 1.8
Assume that \(\left( V_{1}\right) \)–\(\left( V_{2}\right) \) and \((P_1)\)–\(\left( P_{2}\right) \) are satisfied. For each \(\epsilon >0\), \(x_0\) is chosen such that \(P(x_0)=\sup _{x\in {\mathbb {R}}^3}P(x)\), we consider the following function
which is a solution of the minimization problem \(S=\inf \{\Vert \nabla v\Vert ^2_2:v\in {\mathcal {D}}^{1,2}({\mathbb {R}}^3),\Vert v\Vert _6=1\}\), and \(\omega _{\epsilon ,x_0}\) satisfies
Let \(\eta \in {\mathcal {C}}^\infty _0({\mathbb {R}}^3,[0,1])\) be a cutoff function such that \(\eta (x)=1\) for \(|x-x_0|\le r\), \(0<\eta (x)<1\) for \(r<|x-x_0|\le 2r\) and \(\eta (x)=0\) for \(|x-x_0|\ge 2r\). Set \(v_\epsilon (x)=\eta (x)\omega _{\epsilon ,x_0}(x)\), then as \(\epsilon \rightarrow 0\), the following estimations hold (see [9, 25])
As \(I(0)=0\), \(\lim _{t\rightarrow +\infty }I(tv_\epsilon )=-\infty \), there exist \(t_\epsilon >0\) and \(M>0\) such that
Hence, we get
which shows that there exists a constant \(A_1>0\) such that \(t_\epsilon \ge A_1\).
Moreover,
which gives that there exists a constant \(A_2>0\) such that \(t_\epsilon \le A_2\).
Next, we claim that
Choose \(\delta >0\) small enough such that \(|x-x_0|<\delta \epsilon ^{\frac{1}{2}}<r\). Using the condition \((P_2)\), there exists \(\gamma >0\) such that
hence
Then, we have that
which implies the claim holds.
Thus,
If \(4<p<6\) with \(\mu >0\) or \(p=4\) with \(\mu \) sufficient large, we have \(A<0\).
Combine Lemmas 5.2 and 5.4, we can easily verity that if there exists \(v_0\in E\) and \(v_0\not \equiv 0\) such that
Then, the problem (1.6) has at least one positive solution. Taking \(v_0=v_\epsilon \) for \(\epsilon \) small enough, then (5.12) holds, this completes our proof. \(\square \)
5.2 The Case of \(p\in (2,4)\)
Inspired by [12], we consider the truncated functional
where \(\chi (s)=1\) for \(0\le s\le 1\), \(\chi (s)\in (0,1)\) for \(1\le s\le 2\), \(\chi (s)=0\) for \(2\le s\) and \(\Vert \chi '\Vert _\infty \le 2\).
Lemma 5.6
The functional \(I_T\) satisfies the mountain pass geometry:
-
(i)
there exist \(\alpha ,\rho >0\) such that \(I(u)\ge \alpha \) for all \(\Vert u\Vert =\rho \);
-
(ii)
there exists \(e\in H^1(\mathbb (R)^3){\setminus }\{0\}\) such that \(I(e)<0\) with \(\Vert e\Vert >\rho \).
Proof
As \(q>0\), we have
Then, there exists \(\rho >0\) small enough such that \(I_T(u)>0\) as \(\Vert u\Vert =\rho \).
In addition, consider \(u_\tau =\tau u\), we get
and there exists \(\tau _{0}>0\) large enough such that \(I_{\lambda }\left( \tau _{0} u\right) <0\). By taking \(v=u_{\tau _o}\), we have \(I_T{}(v)<0\).\(\square \)
By mountain pass theorem, there exists a sequence \(\{u_n\}\) in E at the level \(c_T\), such that
where
Remark 5.7
It follows from \(I_T\le I\) that \(c_T\le c\), where the constant c is given in (5.1).
Lemma 5.8
There exist \(T_0>0\) independent of q and \(q_*>0\) dependent on \(T_0\) such that if \(q<q_*\), then the sequence \(\{u_n\}\) in (5.13) satisfies
Proof
Similar to the proof in [12]. Arguing by contradiction, suppose for any \(T>0\), there exists \(q>0\) such that
Notice that
Let e be as in Lemma 5.6. As \(n\rightarrow +\infty \), we have
If \(t^2\Vert e\Vert ^2>2T^2\) then \(J_1=0\) and if \(t^2\Vert e\Vert ^2\le 2T^2\), then
By Theorem 1.2, we have
Through similar discussions, we obtain
and
Combining (5.14)–(5.18), we get
Obviously, the inequality (5.19) does not hold if q small enough and for large T. \(\square \)
Remark 5.9
Since \(\limsup \limits _n\Vert u_n\Vert <T_0\), we have \(I_{T_0}(u_n)=I(u_n)\).
Proof of Theorem 1.9
Assume that \(\left( V_{1}\right) \)–\(\left( V_{2}\right) \) and \((P_1)\)–\(\left( P_{2}\right) \) are satisfied. From Lemmas 5.6, 5.8 and Remark 5.9, there exists \(q_*>0\) such that if \(q\in (0,q_*)\) then I possesses a bounded (PS) sequence \(\{u_n\}\) at level \(c_{T_0}\). Moreover, we already know that if \(c_{T_0}<\frac{1}{3}S^{\frac{3}{2}}\Vert P(x)\Vert _\infty ^{-\frac{1}{2}}\), then \(\{u_n\}\) is compact in E. It remains to show that there exists \(v_0\in E\) and \(v_0\not \equiv 0\) such that
for \(p\in (2,4)\) with \(\mu \) sufficiently large. As its proof is similar with the Proof of Theorem 1.8, we omit it here.\(\square \)
References
Abdolrazaghi, F., Razani, A.: On the weak solutions of an overdetermined system of nonlinear fractional partial integro-differential equations. Miskolc Math. Notes 20, 3–16 (2019)
Bonheure, D., Castéras, J.-B., Moreira dos Santos, E., Nascimento, R.: Orbitally stable standing waves of a mixed dispersion nonlinear Schrödinger equation. SIAM J. Math. Anal. 50, 5027–5071 (2018)
Bopp, F.: Eine lineare theorie des elektrons. Ann. Phys. 430, 345–384 (1940)
Born, M.: Modified field equations with a finite radius of the electron. Nature 132, 282 (1933)
Born, M.: On the quantum theory of the electromagnetic field. Proc. R. Soc. Lond. Ser. 143, 410–437 (1934)
Born, M., Infeld, L.: Foundations of the new field theory. Nature 132, 1004 (1933)
Born, M., Infeld, L.: Foundations of the new field theory. Proc. R. Soc. Lond. Ser. A Math. Phys. Eng. Sci. 144, 425–451 (1934)
Brezis, H., Lieb, E.: A relation between pointwise convergence of function and convergence of functional. Proc. Am. Math. Soc. 88, 486–490 (1983)
Brezis, H., Nirenberg, L.: Positive solutions of nonlinear elliptic equations involving critical Sobolev exponents. Commun. Pure Appl. Math. 36, 437–477 (1983)
Chen, S., Tang, X.: On the critical Schrödinger-Bopp-Podolsky system with general non-linearities. Nonlinear Anal. 195, 111734 (2020)
Cheng, B., Li, G., Tang, X.: Nehari-type ground state solutions for Kirchhoff type problems in \({\mathbb{R}}^N\). Appl. Anal. 98, 1255–1266 (2019)
d’Avenia, P., Siciliano, G.: Nonlinear Schrödinger equation in the Bopp-Podolsky electrodynamics: solutions in the electrostatic case. J. Differ. Equ. 267, 1025–1065 (2019)
Deng, Y., Shuai, S.: Wei, Positive solutions for quasilinear Schrödinger equations with critical growth and potential vanishing at infinity. Commun. Pure Appl. Anal. 13, 2273–2287 (2014)
Fibich, G., Ilan, B., Papanicolaou, G.: Self-focusing with fourth-order dispersion. SIAM J. Appl. Math. 62, 1437–1462 (2002)
Gaetano, S., Kaye, S.: The fibering method approach for a non-linear Schrödinger equation coupled with the electromagnetic field. Publ. Mat. 64, 373–390 (2020)
Jakubowski, T., Sydor, S.: Schrödinger perturbations with concave control function. Filomat 33, 6401–6410 (2019)
Jeanjean, L.: On the existence of bounded Palais-Smale sequences and application to a Landesman-Lazer type problem set on \({\mathbb{R}}^N\). Proc. R. Soc. Edinburgh Sect. A 129, 787–809 (1999)
Lieb, E.H.: Sharp constants in the Hardy-Littlewood-Sobolev inequalities. Ann. Math. 118, 349–374 (1983)
Mie, G.: Grundlagen einer theorie der materie. Ann. Phys. 345, 1–66 (1913)
Podolsky, B.: A generalized electrodynamics. Phys. Rev. 62, 68–71 (1942)
Polidoro, S., Ragusa, M.A.: Harnack inequality for hypoelliptic ultraparabolic equations with a singular lower order term. Rev. Mat. Iberoam. 24, 1011–1046 (2008)
Rabinowitz, P.: Minimax methods in critical point theory with applications to differential equations. In: CBMS Regional Conference Series in Mathematics, vol. 65. American Mathematical Society, Providence (1986)
Tang, X.: Infinitely many solutions for semilinear Schrödinger equations with sign-changing potential and nonlinearity. J. Math. Anal. Appl. 401, 407–415 (2013)
Willem, M., Theorems, M.: Progress in Nonlinear Differential Equations and their Applications. Birkhäuser, Boston (1996)
Zhao, L., Zhao, F.: Positive solutions for Schrödinger–Poisson equations with a critical exponent. Nonlinear Anal. 70, 2150–2164 (2009)
Acknowledgements
This work was supported by the National Natural Science Foundation of China (Grant Nos. 11661053, 11771198, 11901276 and 11961045) the Provincial Natural Science Foundation of Jiangxi, China (20181BAB201003 and 20202BAB211004).
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Maria Alessandra Ragusa.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Zhu, Y., Chen, C. & Chen, J. The Schrödinger–Bopp–Podolsky Equation Under the Effect of Nonlinearities. Bull. Malays. Math. Sci. Soc. 44, 953–980 (2021). https://doi.org/10.1007/s40840-020-00994-9
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40840-020-00994-9