
\documentclass[reqno]{amsart} 

\AtBeginDocument{{\noindent\small
{\em Electronic Journal of Differential Equations},
Vol. 2004(2004), No. 26, pp. 1--5.\newline
ISSN: 1072-6691. URL: http://ejde.math.txstate.edu or 
http://ejde.math.unt.edu
\newline ftp ejde.math.txstate.edu  (login: ftp)}
\thanks{\copyright 2004 Texas State University - San Marcos.}
\vspace{9mm}}

\begin{document}

\title[\hfilneg EJDE-2004/26\hfil Schr\"odinger--Poisson system]
{A note on a 3-dimensional stationary Schr\"odinger--Poisson system}

\author[Khalid Benmlih\hfil EJDE-2004/26\hfilneg]
{Khalid Benmlih}

\address{Khalid Benmlih \hfill\break\indent
Department of Economic Sciences,
University of Fez  \hfill\break\indent
P.O. Box 42A, Fez, Morocco}
\email{kbenmlih@hotmail.com}

\date{}
\thanks{Submitted July 29, 2003. Published February 24, 2004.}
\subjclass[2000]{35J50, 35Q40}
\keywords{Schr\"odinger equation, Poisson equation, standing wave,
\hfill\break\indent
variational method}


\begin{abstract}
  In a previous paper we have proved existence of a ground state for a
  stationary  Schr\"odinger--Poisson system in the whole space 
  $\mathbb{R}^3$ under appropriate assumptions on the data, namely the 
  dopant-density $n^*$ and the effective potential $\widetilde V$. 
  In this note we show that the same result remains true under less 
  restrictive hypotheses.
\end{abstract}

\maketitle
\numberwithin{equation}{section}
\newtheorem{theorem}{Theorem}[section]
\newtheorem{lemma}[theorem]{Lemma}
\newtheorem{definition}[theorem]{Definition}

\section{Introduction}

We are concerned with existence of standing waves (i.e. solutions of the
form $u(t,x)= e^{i\omega t} u(x)$ with a real constant $\omega$) for a
time-dependent Schr\"odinger equation where the electric potential $V$
satisfies a linear Poisson equation. This leads to solving the
stationary Schr\"odinger--Poisson system
\begin{gather}
-\frac 12 \Delta u  +  (V + \widetilde V )u   +  \omega  u
= 0 \quad \mbox{in }   \mathbb{R}^3 \label{1.1}   \\
-\Delta V = | u |^2  -  n^*  \quad \mbox{in }  \mathbb{R}^3  \label{1.2}
\end{gather}
where the dopant-density $n^*$ and the effective potential $\widetilde V $
are given reals functions. An existence result of a solution for
\eqref{1.1}--\eqref{1.2} has been established by Lions \cite{lions} in the 
particular case
where $\widetilde V (x)= -2 /|x|$ and $n^* \equiv 0$, by Nier \cite{nier} 
under
some assumptions on the data essentially when $ \|\widetilde V \|_{L^2}$ and
$ \|n^* \|_{L^2}$ are small enough and also recently by the author 
\cite{benmlih} under
appropriate assumptions on $\widetilde V $ and $ n^* $.

In this note, we show existence of a ground state of 
\eqref{1.1}--\eqref{1.2}
as in \cite{benmlih} but under less restrictive assumptions. More precisely, 
an adequate
modification on the proof of the main result in [1, theorem 1.3] allows us 
to
avoid the condition where $n^* \in L^1 (\mathbb{R}^3)$.

Let us recall firstly the principal theorem and the several steps of its
proof given in \cite{benmlih}: after solving explicitly the Poisson equation 
for any
fixed $u\in H^1 (\mathbb{R}^3)$, we substitute the unique solution then
obtained $V = V(u)$ in the Schr\"odinger equation \eqref{1.1} and show 
existence
of a ground state of
\begin{equation}
-\frac 12 \Delta u  +  ( V(u) + \widetilde V )  u   +  \omega u =
0 \quad \mbox{in }  \mathbb{R}^3  .\label{1.3}
\end{equation}

To this end, we show that the energy functional corresponding to \eqref{1.3} 
is
exactly the expression
\begin{equation}\label{functional}
E(\varphi) := \frac 1{4}\int_{\mathbb{R}^3}| \nabla \varphi |^2 dx+\frac
1{4}\int_{\mathbb{R}^3}|\nabla V(\varphi)|^2 dx+\frac 1{
2}\int_{\mathbb{R}^3} \widetilde V \varphi^2
dx+\frac\omega{2}\int_{\mathbb{R}^3} \varphi^2 dx
\end{equation}
and a solution of \eqref{1.3} is obtained as a minimizer of $E$ on 
$H^1(\mathbb{R}^3)$.

Before giving the assumptions imposed on $\widetilde V $ and $ n^* $ to
solve the system \eqref{1.1}-\eqref{1.2}, we recall the following concepts.

\begin{definition} \label{def1.1} \rm
We say that $g$ satisfies the decomposition \eqref{1.5} if:
\begin{itemize}
\item[(i)] $g \in L^1_{\rm loc} (\mathbb{R}^3)$,
\item[(ii)] $g \geq 0$, and
\item[(iii)] There exists $q_0 \in [3/2 , \infty ]$ such that for all 
$\lambda > 0$
there exists $g_{1\lambda} \in L^{q_0}(\mathbb{R}^3)$,
$q_{\lambda} \in ]3/2 , \infty [$ and
$g_{2\lambda} \in L^{q_\lambda} (\mathbb{R}^3)$ such that
\begin{equation}
g = g_{1\lambda} + g_{2\lambda} \quad \mbox{and} \quad
\lim_{\lambda \to 0} \|g_{1\lambda} \|_{L^{q_0}} = 0.\label{1.5}
\end{equation}
\end{itemize}
\end{definition}

As interesting examples of this definition we may consider
$g(x)= 1/|x|^\alpha$  for some $ 0< \alpha < 2 $ or $g\in L^r(\mathbb{R}^3)$
for some $ r >  3/2$ (taking $|g|$ if $g$ is negative).

In what follows we will denote by $\|\cdot \|$  the norm $\|\cdot \|_{L^2}$ 
on
$L^2 (\mathbb{R}^3)$ and by $ [ E \leq c ]$ the set $\{ \varphi ; E(\varphi) 
\leq c\}$.

Consider now the following hypotheses:
\begin{gather}
{\widetilde V}^+ \in L^1_{\rm loc} (\mathbb{R}^3)\quad  \mbox{and} \quad
{\widetilde V}^- \mbox {satisfies the decomposition } \eqref{1.5} \label{H1}
\\
n^* \in L^1 \cap L^{6/5} (\mathbb{R}^3) \label{H2} \\
\inf \Big\{ \int_{\mathbb{R}^3} \left(|\nabla \varphi |^2 +  \varrho (x)
\varphi^2 \right)dx  ,  \int_{\mathbb{R}^3} |\varphi |^2 = 1  \Big\}
< 0 \label{H3}
\end{gather}
where $\displaystyle\varrho (x):= 2 \widetilde V (x) -  { \frac 1{2\pi}
\int_{\mathbb{R}^3} \frac {n^*(y)}{|x-y|} dy } $.

The main result in \cite{benmlih} is as follows.

\begin{theorem} \label{thm1.2}
Assuming \eqref {H1}, \eqref{H2} and \eqref{H3} there exists $\omega_* > 0$
such that for all $0 <\omega < \omega_*$ the equation \eqref{1.3} has a
nonnegative solution $u\not\equiv 0$ which minimizes the functional $ E $
given by \eqref{functional}:
$$ E(u) = \min_{ \varphi \in H^1 (\mathbb{R}^3)} E (\varphi ) .$$
\end{theorem}

The proof of this theorem is divided into the four following Lemmas.

\begin{lemma} \label{lem1.3}
Let $\omega \geq 0 $ and $c \in \mathbb{R} $. If the set $ [ E \leq c ] $
is bounded in $L^2 (\mathbb{R}^3)$ then it is also bounded in $H^1
(\mathbb{R}^3)$.
\end{lemma}

\begin{lemma} \label{lem1.4}
For all $\omega > 0 $ and $c \in \mathbb{R} $ the set $ [ E \leq c ] $ is
bounded in $L^2 (\mathbb{R}^3)$.
\end{lemma}

\begin{lemma} \label{lem1.5}
For any $ \omega > 0 $ the functional $ E $ is weakly lower semicontinuous
on $ H^1 (\mathbb{R}^3) $ and attains its minimum on $ H^1 (\mathbb{R}^3)$
at $ u \geq 0$.
\end{lemma}

\begin{lemma} \label{lem1.6}
There exists $ \omega_* > 0 $ such that if $ 0< \omega < \omega_*$ then $ E
(u) < E (0) $ and thus $u\not\equiv 0$.
\end{lemma}

After analyzing the proofs of the four Lemmas above given in \cite{benmlih}, 
we
remark that theorem \ref{thm1.2} remains true even if we replace the
condition \eqref{H2} by
\begin{equation}
n^* \in L^{6/5} (\mathbb{R}^3) .\label{H2'}
\end{equation}
In the sequel we shall minimize the energy functional $E$ on the space
$$
H := \big\{ u\in H^1 (\mathbb{R}^3) : \int_{\mathbb{R}^3} {\widetilde
V}^+ u^2 dx < \infty  \big\}
$$
which is a Hilbert space, continuously embedded in $H^1 (\mathbb{R}^3)$,
when endowed it with its natural scalar product and norm
$$
(\varphi|\psi ):= \int_{\mathbb{R}^3}\left( \nabla\varphi \cdot \nabla\psi
+ \varphi \psi + {\widetilde V}^+ \varphi \psi \right) dx,  \quad
\|\varphi\|_H := (\varphi|\varphi)^{1/2} .
$$
Consequently Theorem \ref{thm1.2} becomes

\begin{theorem} \label{thm1.7}
Assuming \eqref {H1}, \eqref{H3} and \eqref{H2'} there exists $\omega_* > 0$
such that for all $0 <\omega <\omega_*$ the equation \eqref{1.3} has a
nonnegative solution $u\not\equiv 0$ which minimizes on the space $H $ the
functional $ E $:
$$ E(u) = \min_{ \varphi \in H } E (\varphi ) .$$
\end{theorem}


\section {Preliminaries}

Here we recall the three following Lemmas which will be useful in the
sequel.
\begin{lemma} \label{lem2.1}
Let $n^*\in L^{6/5}(\mathbb{R}^3)$. For all $\varphi\in H^1(\mathbb{R}^3)$ 
the Poisson equation
\begin{equation}\label{2.1}
- \Delta V  =  |\varphi|^2 - n^* \quad \mbox{in } \mathbb{R}^3
\end{equation}
has a unique solution $V:=V(\varphi) \in\mathcal{D}^{1,2} (\mathbb{R}^3)$ 
given
by
\begin{equation}\label{2.2}
V(\varphi) (x)= \frac 1{{4\pi}}\int_{\mathbb{R}^3} \frac {(|\varphi|^2 - n^*
)(y)}{ | x-y |} dy.
\end{equation}
Moreover if we denote by
$$I(\varphi) :=  \frac 1{4} \int_{\mathbb{R}^3} |\nabla V (\varphi) |^2 
dx,$$
then $I$ is $C^1$ on $H^1 (\mathbb{R}^3)$ and its derivative satisfies
$$
\langle I' (\varphi) , \psi \rangle
=  \int_{\mathbb{R}^3} V (\varphi) \varphi \psi dx \quad \forall
\psi \in H^1 (\mathbb{R}^3).
$$
\end{lemma}

For the proof of this lemma see [1, Lemma 2.1, Lemma 2.2].

This Lemma shows in particular that the energy functional corresponding to
\eqref{1.3} is exactly the expression given in \eqref{functional}, namely
$$
E(\varphi) := \frac 1{ 4}\int_{\mathbb{R}^3}| \nabla \varphi |^2
dx+I(\varphi)+\frac 1{ 2}\int_{\mathbb{R}^3} \widetilde V \varphi^2 dx+
\frac \omega {2}\int_{\mathbb{R}^3} \varphi^2 dx .
$$

\begin{lemma} \label{lem2.2}
Let $\theta \in L^r (\mathbb{R}^3)$ for some $ r \geq {3/2}$ then for all
$\delta > 0$ there exists $C_\delta > 0$ such that
$$
\int_{\mathbb{R}^3} \theta (x) |\varphi (x) |^2 dx
\leq  \delta  \|\nabla \varphi \|^2 + C_\delta \| \varphi \|^2
\quad \forall \varphi\in H^1 (\mathbb{R}^3) .
$$
\end{lemma}

For the proof of this lemma see \cite{benmlih} or \cite{brezis}.

Remark that since ${\widetilde V}^-$ satisfies the decomposition \eqref{1.5}
then for any fixed $\lambda > 0$ we have ${\widetilde V}^- = {\widetilde
V}_{1\lambda}^- + {\widetilde V}_{2\lambda}^- $ where for $i=1, 2$,
${\widetilde V}_{i\lambda}^- \in L^s (\mathbb{R}^3)$ for some $s\in [3/2 ,
\infty]$ ($s=q_0$ or $s=q_\lambda$).  Hence taking $\theta:= {\widetilde
V}_{i\lambda}^-$ the inequality of Lemma \ref{lem2.2} holds for $i=1,2$ and
consequently  for all $\delta > 0$ there exists $C_\delta >0 $ so that
\begin{equation}
\int_{\mathbb{R}^3} {\widetilde V}^- (x) |\varphi (x) |^2 dx   \leq  \delta
\| \nabla \varphi \|^2   +  C_\delta \| \varphi \|^2 \quad \forall \varphi
\in H^1 (\mathbb{R}^3) .\label {2.3}
\end{equation}

\begin{lemma} \label{lem2.3}
Let $\psi \in L^r(\mathbb{R}^3)$ for some $ r > {3/2}$. If $v_n
\rightharpoonup 0$ weakly in $H^1 (\mathbb{R}^3)$ then
$$
\int_{\mathbb{R}^3} \psi (x)  v^2_n (x) dx \to  0 \quad  as \quad n \to  
+\infty
$$
\end{lemma}
For the proof of this lemma see [1, Lemma 2.5].

\section {Proof of Theorem \ref{thm1.7}}

We will use once again the same steps as in \cite{benmlih}. Remark at first 
that the
proofs given in \cite{benmlih} for Lemma \ref{lem1.5} and Lemma \ref{lem1.6} 
do not
require the hypothesis $ n^* \in L^1 (\mathbb{R}^3)$ and consequently remain
valid  assuming \eqref{H2'} instead of \eqref{H2}.


\begin{proof}[Proof of Lemma \ref{lem1.3}]
We show here that if the set
$$  [ E \leq c ] := \{ \varphi \in H ; E(\varphi) \leq c\}$$
is bounded in $L^2(\mathbb{R}^3)$ then it is also bounded in $H$.
Indeed since $I(\varphi)$ and $\omega$ are both nonnegative, the inequality
$E(\varphi)\leq c$ gives in particular
$$
\frac 1{ 4}\|\nabla \varphi \|^2  + \frac 1{2}\int\widetilde V^+\varphi^2
dx - \frac 1{2}\int\widetilde V^-\varphi^2 dx  \leq  c .
$$
Now using the estimate \eqref{2.3} with  $\delta =1/4 $ we get
$$
\frac 1{ 8}\|\nabla \varphi \|^2 + \frac 1{2}\int\widetilde V^+\varphi^2
dx  \leq  K_0 \|\varphi \|^2  + c  .
$$
for some constant $K_0 > 0$. \end{proof}

Let us recall that in \cite{benmlih}  we have decomposed the expression of 
$E(\varphi)$
as
$$ E(\varphi)= E_1(\varphi) -  E_2(\varphi) +  E_3(\varphi) +  E(0) $$
where
\begin{gather*}
E_1(\varphi):=  \frac 1{ 4}\int|\nabla\varphi|^2 \,dx + \frac 1{2}
\int\widetilde V^+\varphi^2 dx + \frac{\omega}{2}\int \varphi^2 \,dx \\
  E_2(\varphi):=  \frac 1{2} \int \widetilde V^- \varphi^2 \, dx
  + \frac 1{ 8 \pi}\iint\frac{n^* (y)}{|x-y|}  \varphi^2 (x) \,dx\,
dy\\
E_3(\varphi):= \frac 1{16\pi} 
\iint\frac{\varphi^2(x)\varphi^2(y)}{|x-y|}\,dx \, dy \\
E(0):=  \frac 1{16\pi} \iint\frac{n^*(x) n^*(y)}{|x-y|}\,dx\, dy .
\end{gather*}
Indeed, for the term $I(\varphi)$ it suffices to multiply the equation
\eqref{2.1} by $V(\varphi)$, integrate by parts and use the formula
\eqref{2.2}.

In the proof of the similar lemma [1, Lemma 3.1] we have estimated 
$E_2(\varphi)$ instead
of $\int\widetilde V^-\varphi^2 dx$. More precisely we have estimated the
second term of  $E_2(\varphi)$ by using a certain inequality of type {\sl
Hardy} and the fact that $n^* \in L^1 (\mathbb{R}^3)$.

We point out finally that the decomposition of $E(\varphi)$ as above remains
useful for the rest of proofs.

\begin{proof}[Proof of Lemma \ref{lem1.4}]
Assume by contradiction that there
exists a sequence $(u_j)_j \subset H$ such that $E(u_j)
\leq c$ and $\| u_j \| \longrightarrow +\infty$ as $j\rightarrow +\infty$.
Let $v_j := u_j/\| u_j \|$ then  $\| v_j \| = 1$
and from $E(u_j) \leq c$ we get
\begin{equation}
\frac 1{ 4} \int |\nabla v_j |^2  dx -   E_2 (v_j)  +  E_3 (v_j) \|u_j\|^2
+   \frac\omega{2}  \leq  \frac {c_0} {\| u_j \|^2} \label{3.1}
\end{equation}
where $c_0 = c - E(0)$.
To estimate $E_2 (v_j)$ it suffices to use \eqref{2.3} for the
first term $\int\widetilde V^- v_j^2 dx$ . As to the second, unlike the
proof in \cite{benmlih} we do not require here the assumption
$n^* \in L^1(\mathbb{R}^3)$. Indeed, setting
\begin{equation}
V^*(x):= \iint_{\mathbb{R}^3\times \mathbb{R}^3} \frac {n^* (y)}
{|x-y|} dy = - V(0)(x) \label{3.2}
\end{equation}
as denoted in Lemma \ref{lem2.1} we may write
$$
\iint_{\mathbb{R}^3\times \mathbb{R}^3} \frac {n^* (y)}{|x-y|}  v_j^2
(x) dx dy = \int_{\mathbb{R}^3} V^*(x)  v_j^2 (x) dx .
$$
Knowing that $V(0) \in L^6 (\mathbb{R}^3)$ we can use once more Lemma 
\ref{lem2.2}
with $\theta:=V^*$.

On the whole, we obtain in particular
$$  E_2 (v_j) \leq \frac 1{ 8} \|\nabla v_j \|^2 + K_0
$$
for some positive constant $K_0$ and consequently  we infer from the
inequality \eqref{3.1} that
$$
\frac 1{8} \|\nabla v_j \|^2   +   E_3 (v_j) \|u_j\|^2
+  \frac \omega{2} \leq   \frac {c_0} {\| u_j \|^2}   +  K_0  .
$$
For the remainder of the proof, we conclude exactly as in of [1, Lemma 3.2].
Precisely we show first that, up to a subsequence, $v_j \rightharpoonup 0$
weakly in  $H^1 (\mathbb{R}^3)$. Next, from \eqref{3.1} it follows in
particular that
\begin{equation}
\frac{\omega}{2} -  E_2 (v_j) \leq  \frac{c_0}{\| u_j \|^2}.\label{3.3}
\end{equation}
Using the decomposition ${\widetilde V}^- = {\widetilde V}_{1\lambda}^- +
{\widetilde V}_{2\lambda}^- $ and \eqref{3.2}, we show according to Lemma
\ref{lem2.3} that $E_2 (v_j)\longrightarrow 0$ as $j\rightarrow \infty$.
Finally, letting $j$ go to infinity in \eqref{3.3} we obtain a contradiction
since $\omega$ is positive.
Consequently, all $(u_j)_j\subset H$ such that
$ E(u_j) \leq  c$ is bounded in $L^2
(\mathbb{R}^3)$.
\end{proof}

\begin{thebibliography}{00}

\bibitem{benmlih} Kh. Benmlih, Stationary Solutions for a
Schr\"odinger--Poisson System in $\mathbb{R}^3$; Electron. J. Diff. Eqns.,
Conf. 09 (2002), pp. 65-76. (http://ejde.math.swt.edu)

\bibitem{brezis} H. Brezis \& T. Kato: Remarks on the Schr\"odinger operator
with singular complex potentials; J. Math. Pures \& Appl. N 2, 58 (1979),
137-151.

\bibitem{lions} P. L. Lions: Some remarks on Hartree equation; Nonlinear
Anal., Theory Methods Appl. 5 (1981), 1245-1256.

\bibitem{nier} F. Nier: Schr\"odinger--Poisson systems in dimension
$d\leq 3$, the whole space case; Proceedings of the Royal Society
of Edinburgh, 123A (1993), 1179-1201.

\end{thebibliography}
\end{document}

