Loading [MathJax]/jax/element/mml/optable/Arrows.js
Research article Special Issues

Multiple positive solutions for the logarithmic Schrödinger equation with a Coulomb potential

  • In this article, we mainly study the global existence of multiple positive solutions for the logarithmic Schrödinger equation with a Coulomb type potential

    Δu+V(ϵx)u=λ(Iα|u|p)|u|p1+ulogu2 in R3,

    where uH1(R3), ϵ>0, V is a continuous function with a global minimum, and Coulomb type energies with 0<α<3 and p1. We explore the existence of local positive solutions without the functional having to be a combination of a C1 functional and a convex semicontinuous functional, as is required in the global case.

    Citation: Fangyuan Dong. Multiple positive solutions for the logarithmic Schrödinger equation with a Coulomb potential[J]. Communications in Analysis and Mechanics, 2024, 16(3): 487-508. doi: 10.3934/cam.2024023

    Related Papers:

    [1] Xueqi Sun, Yongqiang Fu, Sihua Liang . Normalized solutions for pseudo-relativistic Schrödinger equations. Communications in Analysis and Mechanics, 2024, 16(1): 217-236. doi: 10.3934/cam.2024010
    [2] Xiulan Wu, Yaxin Zhao, Xiaoxin Yang . On a singular parabolic $ p $-Laplacian equation with logarithmic nonlinearity. Communications in Analysis and Mechanics, 2024, 16(3): 528-553. doi: 10.3934/cam.2024025
    [3] Rui Sun, Weihua Deng . A generalized time fractional Schrödinger equation with signed potential. Communications in Analysis and Mechanics, 2024, 16(2): 262-277. doi: 10.3934/cam.2024012
    [4] Shengbing Deng, Qiaoran Wu . Existence of normalized solutions for the Schrödinger equation. Communications in Analysis and Mechanics, 2023, 15(3): 575-585. doi: 10.3934/cam.2023028
    [5] Maomao Wu, Haidong Liu . Multiple solutions for quasi-linear elliptic equations with Berestycki-Lions type nonlinearity. Communications in Analysis and Mechanics, 2024, 16(2): 334-344. doi: 10.3934/cam.2024016
    [6] Yan-Fei Yang, Chun-Lei Tang . Positive and sign-changing solutions for Kirchhoff equations with indefinite potential. Communications in Analysis and Mechanics, 2025, 17(1): 159-187. doi: 10.3934/cam.2025008
    [7] Zhen Wang, Luhan Sun . The Allen-Cahn equation with a time Caputo-Hadamard derivative: Mathematical and Numerical Analysis. Communications in Analysis and Mechanics, 2023, 15(4): 611-637. doi: 10.3934/cam.2023031
    [8] Yonghang Chang, Menglan Liao . Nonexistence of asymptotically free solutions for nonlinear Schrödinger system. Communications in Analysis and Mechanics, 2024, 16(2): 293-306. doi: 10.3934/cam.2024014
    [9] Xiao Qing Huang, Jia Feng Liao . Existence and asymptotic behavior for ground state sign-changing solutions of fractional Schrödinger-Poisson system with steep potential well. Communications in Analysis and Mechanics, 2024, 16(2): 307-333. doi: 10.3934/cam.2024015
    [10] Yangyu Ni, Jijiang Sun, Jianhua Chen . Multiplicity and concentration of normalized solutions for a Kirchhoff type problem with $ L^2 $-subcritical nonlinearities. Communications in Analysis and Mechanics, 2024, 16(3): 633-654. doi: 10.3934/cam.2024029
  • In this article, we mainly study the global existence of multiple positive solutions for the logarithmic Schrödinger equation with a Coulomb type potential

    Δu+V(ϵx)u=λ(Iα|u|p)|u|p1+ulogu2 in R3,

    where uH1(R3), ϵ>0, V is a continuous function with a global minimum, and Coulomb type energies with 0<α<3 and p1. We explore the existence of local positive solutions without the functional having to be a combination of a C1 functional and a convex semicontinuous functional, as is required in the global case.



    Recently, some studies have focused on the nonlinear Schrödinger equation

    iϵtΨ=ϵ2ΔΨ+(V(x)+w)Ψλ(Iα|Ψ|p)|Ψ|p1Ψlog|Ψ|2, (1.1)

    where Ψ:[0,)×RNC,N3, α(0,N), p>1, λ is a physical constant and Iα is the Riesz potential, defined for xRN{0} as

    Iα(x)=Aα|x|Nα,  Aα=Γ(Nα2)Γ(α2)πN/22α.

    The problem described in equation (1.1) has various practical applications in fields such as quantum mechanics, quantum optics, nuclear physics, transport and diffusion phenomena, open quantum systems, effective quantum gravity, theory of superfluidity, and Bose-Einstein condensation. Notably, periodic potentials V can play a significant role in crystals and artificial crystals formed by light beams. While the logarithmic Schrödinger equation has been excluded as a fundamental quantum wave equation based on precise neutron diffraction experiments, there is ongoing discussion regarding its suitability as a simplified model for certain physical phenomena. The existence and uniqueness of solutions for the associated Cauchy problem have been investigated in an appropriate functional framework [1,2,3], and orbital stability of the ground state solution with respect to radial perturbations has also been studied [4,5,6]. The results regarding the wave equation can be referred to in [7,8,9,10].

    In the Schrödinger equation, the convolution term involve the Coulomb interaction between electrons or interactions between other particles. In Schrödinger equations with convolution terms, this term typically represents the potential energy arising from interactions between particles. Physically, it implies that particles are influenced not only by external potential fields but also by the potential fields created by other particles. These interactions could involve electromagnetic forces, gravitational forces, or other types of interactions depending on the nature of the system. The introduction of the convolution term adds complexity to the Schrödinger equation because particle interactions are often non-local, extending across the entire spatial domain[11]. Overall, Schrödinger equations with convolution terms provide a more realistic description of interactions in multi-particle systems, enabling a more accurate understanding and prediction of the behavior of microscopic particles under mutual influences.

    Understanding the solutions of the elliptic equation

    Δu+V(ϵx)u=λ(Iα|u|p)|u|p1+ulogu2 in RN (1.2)

    holds significant significance in the examination of standing wave solutions for equation (1.1). These standing wave solutions, characterized by the form Φ(t,x)=eiwt/ϵu(x), play a crucial role in various contexts and provide valuable insights into the behavior and properties of the equation.

    In 2018, C. O. Alves and Daniel C. de Morais Filho [12] focus on investigating the existence and concentration of positive solutions for a logarithmic elliptic equation

    {ϵ2Δu+V(x)u=ulogu2, in RN,uH1(RN),

    where ϵ>0, N3 and V is a continuous function with a global minimum. To study the problem, the authors utilize a variational method developed by Szulkin for functionals that are a sum of a C1 functional with a convex lower semicontinuous functional.

    In 2020, Alves and Ji [13] investigated the existence of multiple positive solutions for a logarithmic Schrödinger equation

    {ϵ2Δu+V(x)u=ulogu2, in RN,uH1(RN),

    where ϵ>0, N1 and V is a continuous function with a global minimum. By employing the variational method, the study demonstrates that when the parameter ϵ is sufficiently small, the number of nontrivial solutions is influenced by the "shape" of the graph of the function V.

    In recent years, many authors have studied the nonlinear Schrödinger equation with the potential V. In 2022, Guo et al. [14] utilized fractional logarithmic Sobolev techniques and the linking theorem to elucidate existence theorems for equations with logarithmic nonlinearity. Further, a recent study [15] delineates conditions for a singular nonnegative solution in bounded Rn domains (n2), providing comprehensive insights into its behavior.

    Inspired by the outcomes observed in the aforementioned papers, in this paper we aim to investigate the existence of multiple positive solutions for the problem (1.2) when N=3, λ>0 and 1p2. It is noteworthy that the introduction of a convolution term presents a notable aspect. The difficulty arises in analyzing the unique existence of solutions to the energy functional when both the convolution term and the logarithmic term operate concurrently. Addressing this challenge involves employing specialized analytical techniques, setting it apart from the methods utilized in [13], marking a novel approach.

    In this paper, we shall prove the existence of solution for (1.2) in H1(R3). The associated energy functional of (1.2) will be defined as Jϵ:H1(R3)(,+),

    Jε(u)=12R3(|u|2+(V(εx)+1)u2)dxλ2pR3R3|u(x)|p|u(y)|p|xy|3αdxdyR3H(u)dx, (1.3)

    where

    R3H(u)dx=R3u22dx+u2logu22dx,uR3,

    with

    H(u)=u0slogs2ds=u22+u2logu22,

    and

    L(u)=R3R3|u(x)|p|u(y)|p|xy|3αdxdy.

    Given the infinite character and lack of C1 smoothness of the functional Jε, a new approach is required to find weak solutions since traditional methods are not effective here. In this scenario, the fundamental element of our approach lies in harnessing the groundbreaking minimax method introduced by Szulkin [16]. Furthermore, we will employ the Gagliardo-Nirenberg inequality [17,18], the Brezis-Lieb lemma [19], and other specifically techniques for handling the nonlinear Coulomb potential, culminating in a robust result of strong convergence.

    In our research, the potential V is based on the following assumptions[13]:

    1. V:R3R is a continuous function such that

    lim|x|+V(x)=V.

    with 0<V(x)<V for any xR3.

    2. There are l points z1,,zl in R3 with z1=0 such that

    1=V(zi)=minxR3V(x), for 1il.

    By employing the variational method, we can establish the existence of non-trivial solutions for the logarithmic Schrödinger equation with a Coulomb-type potential when ϵ is sufficiently small (ϵ>0). This outcome is contingent upon the distinctive characteristics of the graph of the function V.

    A positive solution of problem (1.2) means that there exists a positive function uH1(R3){0} satisfy u2logu2<+ and

    R3uv+V(εx)uvdx=λR3(Iα|u|p)|u|p1vdx+R3uvlogu2,  for all  vC0(R3).

    The main result is as follows.

    Suppose that V satisfies 1 and 2. There exists ε>0 such that problem (1.2) has l positive soutions in H1(R3) for ε(0,ε).

    The paper is organized as follows. In Section 2. we present several preliminary results that will be employed in the proofs of our main theorems. In Section 3. we prove the main result which are in the local case. In Section 4. we generalize the local results to the global space.

    Notation: Henceforth, in this paper, unless otherwise specified, we adopt the following notations:

    BR(u) denotes an open ball centered at u with a radius of R>0.

    ● If g is a measurable function, the integral RNg(x)dx will be denoted by g(x)dx.

    C, C1, C2 etc. will denote positive constants of negligible importance with respect to their exact values.

    LR(u) denotes the function L(u) within the ball BR(0).

    p denotes the usual norm of the Lebesgue space Lp(R3), for p[1,+).

    on(1) denotes a real sequence with on(1)0 as n+.

    ● The expression  dxdy denotes RNRN dxdy.

    2=2NN2.

    In this section, we give some results and technical tools used for the main results.

    First, we define the effective domain of J,

    D(Jϵ):={uH1(R3):Jϵ(u)<+}.

    Considering the problem

    Δu+V(0)u=λ(Iα|u|p)|u|p1+ulogu2 in R3, (2.1)

    the corresponding energy functional associated to (2.1) is

    J0(u)=12(|u|2+(V(0)+1)u2)dxλ2p|u(x)|p|u(y)|p|xy|3αdxdy12u2logu2dx.

    And define the Nehari manifold

    Σ0={uD(J0)(0):J0(u)u=0},

    where

    D(J0)={uH1(R3):J0(u)<+}.

    The problem (2.1) has a positive solution attained at the infimum,

    c0:=infuΣ0J0(u),

    which will be proved in the Lemma 3. We shall additionally utilize the energy level

    c:=infuΣJ(u),

    through replacing V(0) by V, and

    Σ={uD(J)(0):J(u)u=0},

    it is clear that

    c0<c.

    Regarding to the values of c0 and c, it should be noted that they correspond to the critical levels of the functionals J0 and J, commonly referred to as the Mountain Pass levels.

    Based on the approach discussed in previous studies [12,20,21], we address the issue of J0 and J lacking smoothness by decomposing them into a sum of a differentiable C1 functional and a convex lower semicontinuous functional, respectively. Following by [13], to facilitate this decomposition, for δ>0, we define the following functions:

    F1(s)={0,s=0,12s2logs2,0<|s|<δ,12s2(logδ2+3)+2δ|s|12δ2,|s|δ,

    and

    F2(s)={0,|s|<δ,12s2log(s2/δ2)+2δ|s|32s212δ2,|s|δ.

    Therefore

    F2(s)F1(s)=12s2logs2,sR. (2.2)

    The functionals J0,J:H1(R3)(,+] can be reformulated as an alternative form denoted by

    J0(u)=Φ0(u)+Ψ(u) and J(u)=Φ(u)+Ψ(u),uH1(R3) (2.3)

    where

    Φ0(u)=12(|u|2+(V(0)+1)|u|2)dxλ2pL(u)F2(u)dx (2.4)
    Φ(u)=12(|u|2+(V+1)|u|2)dxλ2pL(u)F2(u)dx (2.5)

    and

    Ψ(u)=F1(u)dx. (2.6)

    The properties of F1 and F2, as demonstrated in [20] and [21], can be summarized as follows:

    F1,F2C1(R,R). (2.7)

    For δ>0 small enough, F1 is convex, even, F1(s)0 for all sR and

    F1(s)s0,   sR. (2.8)

    For each fixed q(2,2), there is C>0 such that

    |F2(s)|C|s|q1,sR. (2.9)

    Utilizing the information provided earlier, it can be deduced that the functional Ψ possesses the properties of convexity and lower semicontinuity. Additionally, we can observe that the function Φ belongs to the class of C1 functions.

    As we've discussed earlier, solutions to equation (1.2) within a localized context can be addressed through conventional techniques. However, the situation undergoes a transformation when we expand our scope to encompass the entire space. Within this broader perspective, it becomes apparent that the functional Ψ lacks the characteristic of continuous differentiability (C1). This particular case necessitates the application of a novel and separate critical point theorem. In the subsequent section, dedicated to the global case, it becomes essential to introduce definitions that were originally presented in the work referenced as [16].

    Let J be a C1 functional defined on Banach space X, we say that {un} is a Palais-Smale sequence of J at c ((PS)c sequence, for short) if

    J(un)c, and J(un)0, as n+ (2.10)

    Let E be a Banach space, E be the dual space of E and , be the duality paring between E and E. Let J:ER be a functional of the form J(u)=Φ(u)+Ψ(u), where ΦC1(E,R) and Ψ is convex and lower semicontinuous. Let us list some definitions:

    1. The sub-differential J(u) of the functional J at a point uH1(RN) is the following set

    {wE:Φ(u),vu+Ψ(v)Ψ(u)w,vu,vE} (2.11)

    2. A critical point of J is a point uE such that J(u)<+ and 0J(u), i.e.,

    Φ(u),vu+Ψ(v)Ψ(u)0,vE (2.12)

    3. A PS sequence at level d for J is a sequence (un)E such that J(un)d and there is a numerical sequence τn0+ with

    Φ(un),vun+Ψ(v)Ψ(un)τnvun,vE (2.13)

    4. The functional J satisfies the PS condition at level d ((PS)d condition, for short) if all PS sequences at level d has a convergent subsequence.

    As [21] Lemma 2.2, J is of class C1 in H1(Ω) with Ω is a bounded domian. Hence we can construct the mountain pass structure and find the boundedness of the (PS) sequence without using the decomposition method in the local case, which is different from [12,13,20,21].

    In order to make the subsequent theorem proof involving the whole space situation clearer, we explain some necessary concepts here. Henceforward, for every ωD(J0), the functional J10(w):H1c(R3)R given by

    J0(w),z=ΦV(w),z+F1(w)z,  zH1c(R3)

    and

    J0(w)=sup{J0(w),z:zH1c(R3),  and   zv1}.

    If J0(ω) is finite, then J0(w) can be extended to a bounded operator in H1(R3) and can be therefore be viewed as an element of (H1(R3)).

    If {un}D(J){0} is a (PS) sequence for Jε, then Jϵ(un)un=on(1)unV. If {un} is bounded, we have

    Jϵ(un)=Jϵ(un)12Jϵ(un)un+on(1)unV=12|un|2dx+λ2(11p)L(un)+on(1)unV,nN.

    In this section, we provide the proof of the existence of l nontrivial critical points for Jϵ,R to equation (1.2) on a local case, which constitutes the preliminary step necessary for our main result. This serves as the foundational work leading up to our primary outcome.

    Fix R0>0 such that ziBR0(0) for all i{1,,l}. So for all R>R0 and uH1(BR(0)),

    Jϵ,R(u)=12BR(0)(|u|2+(V(ϵx)+1)u2)dxλ2pLR(u)12BR(0)u2logu2dx.

    For any u,vH1(BR(0)), it is easy to verify that Jϵ,RC1(H1(BR(0)),R) and

    Jϵ,R(u)v=BR(0)uvdx+V(ϵx)uvdxλBR(0)(Iα|u|p)|u|p1vdxBR(0)uvlogu2dx.

    The local space H1(BR(0)) is endow with the norm

    uV=(BR(0)(|u|2+(V(ϵx)+1)u2)dx)12

    which is also a norm in H1(R3).

    According to the definition of V-norm and H1-norm, we have the following inequality

    C1uH1((|u|2+(V(ϵx)+1)u2)dxλL(u)12)12uVC2uH1.

    One can see that V-norm is equivalent to H1-norm.

    In the subsequent analysis, we denote Σϵ,R as the Nehari manifold correspond to Jϵ,R, which can be defined as follows:

    Σϵ,R={uH1(B){0},Jϵ,R(u)u=0}={uH1(B){0},Jϵ,R(u)=12BR(0)u2+λ2(11p)LR(u)}.

    For all ϵ>0, R>R0, Jϵ,R has the Mountain Pass geometry.

    Proof. (i) Recall that

    Jϵ,R(u)=12BR(0)(|u|2+(V(ϵx)+1)u2)dxλ2pLR(u)12BR(0)u2logu2dx. (3.1)

    Following by the Hardy-Littlewood-Sobolev inequality and Sobolev imbedding, we obtain

    LB(u)|u(x)|p|u(y)|p|xy|Nαdxdy(|u|2NpN+αdx)N+αNCu2pV, (3.2)

    where N+αN<p<N+αN2. And for q>2 small and u>0, we have

    u2logu2dxCq|u|quqV. (3.3)

    Hence, by (3.1),(3.2) and (3.3), it follows that

    Jϵ,R(u)12u2VλC1u2pVC2uqV>C>0,

    for a constant C>0, and uV>0 small enough.

    (ii) Fix uD(J){0} with suppuBR(0), and for s>0, λ>0, we have

    Jϵ,R(su)=12BR(0)(s2|u|2+s2(V(εx)+1)u2)dxλ2ps2pLR(u)12s2logs2BR(0)u2dx12s2BR(0)u2logu2dxs2(12BR(0)(|u|2+(V(εx)+1)u2)dxlogsBR(0)u2dx12BR(0)u2logu2dx).

    Because of the boundness of Jϵ,R, there exist three bounded terms in the right side of the above inequality, except for the third term. Therefore, we obtain that Jϵ,R(u) as s+. So there exists s0>0 independent of ϵ>0 small enough and R>R0 such that Jϵ,R(s0u)<0.

    All (PS) sequence of Jϵ,R are bounded in H1(BR(0)).

    Proof. Let {un}H1(BR(0)) be a (PS)d sequaence. Then,

    |un|2L2(BR(0))+λ(11p)LR(un)2Jϵ,R(un)Jϵ,R(un)un=2d+on(1)+on(1)unVC+on(1)unV. (3.4)

    for some C>0. And we ultilize the following logarithmic Sobolev inequality [11],

    u2logu2a2πu2L2(RN)+(logu2L2(RN)N(1+loga))u2L2(RN) (3.5)

    for all a>0. By taking a2π=12, ξ(0,1) and combining (3.4) and(3.5) we get

    BR(0)u2nlogu2n14un22+C(1+unV)1+ξ. (3.6)

    Above all, for some ξ(0,1),

    d+on(1)=Jϵ,R(un)=12BR(0)|un|2+12BR(0)(V(ϵx)+1)u2nλ2pLR(un)12BR(0)u2nlogu2nCun2V(1+unV)1+ξλ2pLR(un).

    By (3.4) we have λ2pLR(un)λ2(11p)LR(un)C+on(1)unV, α(N2,N);p(2,N+αN2) therefore it implies that

    Cun2VC(1+unV)1+ξ+C+on(1)unV,

    which means unVC, i.e. (un) is bounded in H1(BR(0)).

    Fix u00, u0H1(BR(0)) and u20logu20dx>. According to

    cϵ,R=infγΓsupt[0,1]Jϵ,R(γ(t))supt>0Jϵ,R(tu0)=D0.

    where the definition of the path set γ is given in the lemma 3 and D0 is a uniform constant. Hence we obtain {un} is also bounded in H1(R3).

    Now, for a fixed uD(J0){0}, and t>0. Define the function

    tϕ(t):=Jϵ(tu).

    Via computation, we have

    ϕ(t)=t((|u|2+V(ϵx)u2)dxλt2p2L(u)2logtu2dxu2logu2dx).

    Setting f(t)=λat2p1+2blogt, for a,b>0 and p>1. In the following, we prove that there exists an unique critical point ˜t, with ˜t>0, at which the function ϕ attains its maximum positive value.

    1. According to Mountain Pass Geometry, there exists ˜t>0 such that f(˜t)=0, i.e. ϕ(˜t)=0.

    2. Since f(t)=(2p1)λat2p2+2bt>0, we know that the function f is a monotonically increasing function, and furthermore, this means that ϕ reaches a positive maximum at the unique critical point ˜t.

    Hence, for any uD(Jϵ){0}, the intersection of every path {tu;t>0} forms a set

    Σϵ={uD(Jϵ){0};Jϵ(u)=12u2dx+λ2(11p)L(u)}

    exactly at the unique point ˜tu. Moreover, ˜t=1 if and only if

    uΣϵ (˜t=1 ϕ(˜t)=Jϵ(˜tu)u=Jϵ(u)u=0).

    Based on the energy levels shown above, the following results are obtained. For ϵ0,

    cϵ=infuΣϵJϵ(u). (3.7)

    Proof. Let

    Γ:={γC([0,1],H1(R3)):γ(0)=0,J(γ(1))<0}

    we can define the mountain pass energy level

    c:=infηΓsupt[0,1]J(η(t)).

    Let uΣϵ, we consider Jϵ(t0u)<0 for some t0>0. Then for the continuous path γϵ(t)=tt0u, we have

    infγΓsupt[0,1]Jϵ(γϵ(t))=cϵmaxt[0,1]Jϵ(γϵ(t))maxt0Jϵ(tu)=Jϵ(u).

    Hence

    cϵinfuΣϵJϵ(u). (3.8)

    On the other hand, we will prove that cϵinfuΣϵJϵ(u). Take a (PS) sequence {un}H1(R3) for Jϵ. By Lemma 3, (un) is bounded in H1(R3). We claim un20. By contradiction, if un20, using interpolation, unq0, for any q[2,2). Because |F2(s)|C|s|q1, then

    F2(un)un0,

    and using Hardy-Littlewood-Sobolev inequality again, we obtain L(un)0. Recall that

    un2V+F1(un)undx=Jϵ(un)un+λL(un)+u2ndx+F2(un)undx=on(1)unV+λL(un)+u2ndx+F2(un)undx=on(1), (3.9)

    from where it follows that unV0 and F1(un)un0.

    Since F1 is convex, even and F1(t)F1(0)=0, for all tR, we derive that 0F1(t)F1(t)t for all tR. Hence F1(un)0 in L1(R3). Then Jϵ(un)Jϵ(0)=0, which contradicts to cϵ>0. Our claim is proved. Hence, there are constants b1 and b2 such that

    0<b1un2b2. (3.10)

    Next, let tn(0,1), tnunΣϵ, and recalling that

    Jϵ(tnun)=12|tnun|2dx+λ2(11p)L(tnun)=12t2n|un|2dx+(V(ϵx)+1)u2ndxλ2pt2pnL(un)12t2nlogt2nu2ndx12t2nu2nlogu2ndx. (3.11)

    and

    Jϵ(un)un=(|un|2+V(ϵx)u2n)dxλL(un)u2nlogu2ndx.

    Then we get

    λ(t2p2n1)L(un)+logt2nu2ndx=Jϵ(un)un=on(1)unV.

    According to (3.10) and L(u)0, this equation implies tn1. In addition, by (3.11) and Remark 2 we have

    infuΣϵJϵ(u)Jϵ(tnun)=t2n2u2ndx+λ2(11p)t2pnL(un)t2n(12u2ndx+λ2(11p)L(un))=t2n(Jϵ(un)+on(1)unV).

    Therefore, taking the limit we get

    infuΣϵJϵ(u)cϵ.

    The functional Jϵ,R satisfies the (PS) condition.

    Proof. Take a (PS) sequence {un}H1(BR(0)), it means that

    Jϵ,R(un)d,
    Jϵ,R(un)un=on(1)unV.

    By Lemma 3, we know there exists {un}H1(BR(0)), and a subsequence of un, which still denoted by itself such that unV, i.e.

    unu  in  H1(BR(0)),
    unu  in  Lq(BR(0)),q[1,2),
    unu  a.e. in  BR(0).

    From [13], we set f(t)=tlogt2, F(t)=t0f(s)ds=12(t2logt2t2) for all tR and for p(2,2), there is C>0 such that

    |f(t)|C(1+|t|p1), tR

    and

    |F(t)|C(1+|t|p), tR.

    In addition, by definition of the norm in H1(BR(0)), we get

    unu2V=|(unu)|2dx+(V(ϵx)+1)|unu|2dx,
    Jϵ,R(un)(unu)=un(unu)dx+V(ϵx)un(unu)dxλ(Iα|un|2)|unu|undx(unu)unlogu2ndx=|(unu)|2dx+V(ϵx)|unu|2dxλ(Iα|unu|2)|unu|2dxf(un)|unu|dx=on(1).

    Hence, it is easy to see that

    |(unu)|2dx+V(ϵx)|unu|2dx=λ(Iα|unu|2)|unu|2dx+f(un)(unu)dx+on(1)=on(1).

    It implies that

    unuV0,

    which means the sequence {un} satisfies (PS) condition.

    In fact, Theorem 3 concerns the existence of multiple solutions for equation (1.2) on a ball, which is crucial for the study of the existence of multiple solutions on the entire space as we desire. In order to prove this crucial result, we first present several lemmas. Next, we use the tricks in [13], by constructing l small balls and finding the center of mass, it plays a key role in the proof of the following theorem.

    Fix ρ0>0 so that it satisfies ¯Bρ0(zi)¯Bρ0(zj)=ϕ for ij,i,j{1,,l} and li=1Bρ0(zi)BR0(0). Denote Kρ02=li=1¯Bρ02(zi), and define the functional Qε:H1(R3){0}R3 by

    Qε(u)=χ(εx)g(εx)|u|2dxg(εx)|u|2dx

    where χ:R3R3 is given by χ(x)={x,|x|R0.R0x|x|,|x|>R0. and g:R3R3 is a radial positive continuous function with

    g(zi)=1,  i{1,,l}  and  g(x)0, as  |x|+.

    The next lemma provides a useful way to generate (PS)c sequence associated with Jϵ. There exist α0>0, ϵ0>0, and R0>0 such that ε1(0,ε0) small enough and R1>R0 large enough, if uΣε,R and Jε,R(u)c0+α0, then Qε(u)Kρ02 for any ε(0,ε1) and RR1.

    Proof. We prove this lemma by contradiction. If there is αn0, εn0 and Rn, unΣεn,Rn satisfies

    Jεn,Rn(u)c0+αn,

    but

    Qε(un)Kρ02.

    By definition of c0 and Lemma 3, c0cεn,Rn, it is easy to see that

    c0cεn,RnJεn,Rn(un)c0+αn

    which means JεnRn(un)=cεn,Rn+on(1). Denote the functional Ψεn,Rn:H1(BRn(0))R by

    Ψεn,Rn(u)=Jεn,Rn(u)12BRn(0)|u|2λ2(11p)LR(u).

    It implies that

    Σεn,Rn={uH1(BR(0)){0}:Ψεn,Rn(u)=0}.

    Via computation, we obtain

    Ψεn,Rn(u)u=|u|2λ(p1)L(u)β,    nN,

    where β>0 to guarantee cεn,Rn>0. Without loss of generality, we have the above conditions. We can then proceed to apply the Ekeland Variational Principle from Theorem 8.5 in [22], assuming that

    Jεn,Rn(un),   as   n.

    Now, from Jεn,Rn(un)=12BRn(0)|un|2dx+λ2(11p)LRn(un)c0>0, we have lim infnRn>0. And according to Section 6 in [12], there are two cases:

    1. unu0 in L2(RN), and uH1(RN).

    2. There exists (yn)RN such that vn=un(+yn)v0 in L2(RN), and vH1(RN).

    For case (1), recall that our assumption ε0, χ(0)=0 and g(0)=1

    Qεn(un)=χ(εnx)g(εnx)|un|2dxg(εnx)|un|2dxχ(0)g(0)|un|2dxg(0)|un|2dx=0Kρ02.

    This contradicts to QεnKρ02.

    For case (2), there are two different situations. If |εnyn|+, then J(v)v0. Thus, for s(0,1] such that svΣ,

    2c2J(sv)=2J(sv)J(sv)sv=|sv|2+λ(11p)s2p|v|p(x)|v|p(y)|xy|Nαdxdy|v|2+λ(11p)|v|p(x)|v|p(y)|xy|Nαdxdylim infn+|vn|2+λ(11p)|vn|p(x)|vn|p(y)|xy|Nαdxdy=lim infn+|un|2+λ(11p)|un|p(x)|un|p(y)|xy|Nαdxdy=limn2Jεn,Rn(un)=2c0,

    which contradicts c0<c. If εnyny for some yRN, and some subsequence. In this case, the functional JV:H1(RN)R is given by

    JV(u)=12(|u|2+(V(y)+1)u2)dxλ2p|un|p(x)|un|p(y)|xy|Nαdxdy12u2logu2dx,

    and cV is the moutain pass level of JV. Similar as before,

    cV=infuΣVJV(u),

    where

    ΣV={uD(JV){0}:JV(u)=12u2+λ2(11p)|u|p(x)|u|p(y)|xy|Nαdxdy}.

    If V(y)>1=miniV(xi),i{1,,l}, then

    cV>c0,

    but according to the previous arguments

    cVc0,

    which is a contradiction. So V(y)=1 and y=zi for i{1,,l}.

    Qεn(un)=χ(εnx)g(εnx)|un|2dxg(εnx)|un|2dx=χ(εn(x+yn))g(εn(x+yn))|vn|2dxg(εn(x+yn))|vn|2dxχ(zi)g(zi)|v|2dxg(zi)|v|2dx=ziKρ02.

    This is contrary to our initial hypothesis, and the proof is done.

    In the following, for simplicity, we indicate the following notations.

    Ωiε,R{uΣε,R:|Qε(u)zi|<ρ0},Ωiε,R{uΣε,R:|Qε(u)zi|=ρ0},αiε,RinfuΩiε,RJε,R(u),˜αiε,RinfuΩiε,RJε,R(u).

    For γ(cc08,cc02), there exists ε2(0,ε1) small enough such that

    αiε,R<c0+γ    and   αiε,R<˜αiε,R

    for all ε(0,ε2), and  RR1(ε)>R0.

    Proof. Let uH1(R3) be a ground state solution of J0, that is for uΣ0,

    J0(u)=infuΣ0J0(u)=c0,   and   J0(u)=0.

    For any i{1,,l}, there exists ε1>0 such that

    |Qε(u(ziε))zi|<ρ,ε(0,ε1).

    Fix R>R1=R1(ε) and tε,R>0 such that uiε,R(x)=tε,RφR(x)u(xziε)Σε,R,

    |Qε(uiε,R)zi|<ρ,ε(0,ε1)  and   R>R1,

    and

    Jε,R(uiε,R)c0+α08,    ε(0,ε1),R>R1, (3.12)

    where φR(x)=φ(xR) with φC0(R3), 0φ(x)1 for all xR3,φ(x)=1 for xB12(0) and φ(x)=0 for xBc1(0). So

    uiε,RΩiε,Rε(0,ε2)   and   R>R1.

    Take the infimum for (3.12), thanks to α0<cc02,Jε,Rc0+α0<c+c02, we get

    αiε,R<c0+α04<c0+γ. (3.13)

    Now let cc08<γ<cc02, then the first inequality is done. Next, if uΩiε,R, then there is

    uΣε,R   and   |Qε(u)zi|=ρ0>ρ02,

    hence Qε(u)Kρ02. By Lemma 3, we have

    Jε,R(u)>c0+α0 (3.14)

    for uΩiε,R and ε(0,ε2), RR1. Take the infimum for (3.14) we obtain

    ˜αε1R=infΩε,RJε,R(u)c0+α0,ε(0,ε2),  RR1. (3.15)

    Above all, from (3.13) and (3.15)

    αiε,R<˜αiε,Rforε(0,ε2),andRR1,

    where ε2(0,ε1).

    For ε(0,ε2) small enough and R1=R1(ε)>R0 large enough, there exist at least l nontrival critical points of Jε,R for ε(0,ε0) and RR1. Moreover, all of the solutions are positive.

    Proof. From Lemma 3, for ε(0,ε2) small enough and R1>R0 large enough, there is

    αiε,R<˜αiε,R   for  ε(0,ε)   for  RR1.

    As stated Theorem 2.1 in [23], the inequalities mentioned above enable us to employ Ekeland's variational principle to establish the (PS)αiε,R sequence (uin)Ωiε,R for Jε,R. Following by Lemma 3, since αiε,R<c0+γ, there is ui such that uinui in H1(BR(0)). Then

    uiΩiε,R,Jε,R(ui)=αiε,R,Jε,R(ui)=0.

    Recall that

    ¯Bρ0(zi)¯Bρ0(zj)ϕ,ij,

    and

    Qε(ui)¯Bρ0(zi)(QεKρ2=li=1¯Bρ2(zi)).

    We have uiuj,ij,i,j{1,,l}. If we decrease γ and increase R1 when necessary, we can assume that

    2cε,R<c0+γ.

    for ε(0,ε),RR1. So all of the solutions do not charge sign, and because the function f(u)=ulogu2 is odd, we make them nonnegative. The maximum principle implies that any solution to a given equation or system of equations within the open ball BR(0) will necessarily be positive throughout the entire ball, provided that it is positive on the boundary.

    In this section, we prove the existence of solution for the original equation (1.2).

    For vH1(BRn(0)), uin=uiε,Rn be a solution obtained in Theorem 3.

    BRnuinv+V(εx)uinv=λBRn(Iα|uin|p)|uin|p1vdx+BRnuinlog|uin|2vdx,
    Jε,Rn(uin)=αiε,Rn,nN.

    There exists uiH1(R3) satisfies uinui in H1(R3) and ui0, i{1,,l}.

    Proof. From Lemma 3, we know that {αiε,Rn} is a bounded sequence,

    Jε,Rn(uin)=αiε,Rn<c0+γ

    which implies that {uin} is a bounded sequence. So we can assume that uinui for some uiH1(R3).

    Next, we prove ui0. In the following, we use {un} and {αn} to denote {uin} and {αiε,Rn} for convenience.

    To continue, let us utilize the Concentration Compactness Principle, originally introduced by Lions [13], applied to the following sequence.

    ρn(x):=|un(x)|2un22,xR3.

    This principle guarantees that one and only one of the following statements is true for a subsequence for {ρn}, which we will still refer to as {ρn}:

    (Vanishing) For all K>0, one has:

    limn+supyRNBK(y)ρndx=0; (4.1)

    (Compactness) There exists a sequence {yn} in R3 with the property that for all ε>0, there exists K>0 such that for all nN, one has:

    BK(yn)ρndx1η; (4.2)

    (Dichotomy) There exists {yn}RN, α(0,1), K1>0, Kn+ such that the functions ρ1,n(x)=χBK1(yn)(x)ρn(x) and ρ2,n(x):=χBcKn(yn)(x)ρn(x) satisfy:

    ρ1,ndxα    and   ρ2,ndx1α. (4.3)

    Our goal is to demonstrate that the sequence {ρn} satisfies the Compactness condition, and to achieve this, we will exclude the other two possibilities. By doing so, we will arrive at a contradiction, thus proving the proposition.

    The vanishing case (4.1) can not occur, otherwise we deduce that unp0, and consequently F2(un)un<. By employing the same reasoning as in the previous section, it can be proven that un0 in H1(R3). However, this contradicts the fact that αnc1 for all nN, as stated in Lemma 3.

    The Dichotomy case (4.3) can not occur. Let us assume that the dichotomy case holds, under this assumption, we claim that the sequence {yn} is unbounded. If this were not the case and {yn} were bounded, then in that situation, utilizing the fact that unL2(R3)0, the first convergence in (4.3) would lead to

    BK1(yn)|un|2dx=|un|22R3ρ1,ndxδ,

    for some δ>0 and n large enough. Therefore, taking R>0 such that BK1(yn)BR(0) for all nN, it follows that BR(0)|un|2dxδ, for all n sufficiently large. Becauseun0 in L2(BR(0)), the inequality above is impossible. As a result, {yn} is an unbounded sequence. In the following, denote:

    vn(x):=un(x+yn),xR3.

    Since the boundness of the sequence (vn)H1(R3) and up to subsequence, we may assume that vnv. By the first part of (4.3), v0 holds.

    Claim4.1. F1(v)vL1(R3) and J(v)v0. For ηC0(R3), 0η1, η1 in B1(0) and η0 in B2(0)c, we define ηR:=η(R) and v=ηR(yn)un, we get

    vn(ηRvn)dx+(V(ε(x+yn))+1)v2nηRdx+F1(vn)vnηRdx=F2(vn)vnηRdx+λ(Iα|vn|p)|vn|pηRdx+on(1).

    If we fix R and go to the limit in the above equation when n, we get

    |v|2ηRdx+vηRvdx+(V+1)v2ηRdx+F1(v)vηRdxF2(v)vηRdx+λ(Iα|v|p)|v|pηRdx

    where |ηR|2R, using that F1(t)t0 for all tR, and Fatou's lemma as R+, we obtain

    |v|2dx+(V+1)v2dxλ(Iα|v|p)|v|pdx+F1(v)vdxF2(v)vdx0,

    that is J(v)v0.

    On this account, there exists t(0,1] such that tvΣ, then

    cJ(tv)=t22|v|2dx+λ2(11p)t2pL(v)lim infn+[12|vn|2dx+λ2(11p)L(vn)]lim supn+[12|un|2dx+λ2(11p)L(un)]=lim supn+Jεn,Rn(un)=lim supnαnc0+γ.

    But we have γ<cc0, it is absurd. Hence, there is no dichotomy, and in fact compactness must hold. We make the last requirement to achieve our aim.

    Claim4.2. The sequence of points {yn}R3 in (4.2) is bounded.

    To establish this claim, we employ a proof by contradiction by assuming that the sequence of {yn} is bounded. However, by considering a subsequence, we observe that |yn|+. Following a similar approach as in the case of the Dichotomy, where {yn} was unbounded, we eventually arrive at the inequality c0+γc.

    For a given η>0, there is R>0 such that

    BcR(0)ρndx<η,nN,

    that is

    BcR(0)|un|2dxη|un|22ηsupnN|un|22=bη.

    Therefore, for R1max{R,R}, since un0 in L2(BR1(0)), there is n0N large enough such that

    BR1(0)|un|2dxη,nn0.

    Thereby, we conslude

    |un|2dxη+BcR1(0)|un|2dxη+bηCη,

    where Cη. Due to the arbitrary nature of η, we can deduce that un0 in L2(R3). By interpolation on the Lebesgue spaces and {un} is bounded in H1(R3), it follows that

    un0   in   Lp(R3),2p<2.

    Using the trick that for some p>1 small, tlogtCtp, it implies that

    u2nlogu2n0.

    For p(3+α3,3+α), the sequence {unp}nN converges to up in the sense of measures, {un}nN converges to u almost everywhere, the sequence {Iα|un|p}nNis bounded in L2(R3) and u0.

    From Proposition 4.8 in [24], since unD(J){0} then we have

    limnR3Iα|un|p)|un|p(Iα|unu|p)|unu|p=(Iα|u|p)|u|p. (4.4)

    Above all, Jε,Rn(un)=αn0, which contradicts αncε>0, for all nN.

    Proposition 4 yields a direct corollary as follows. For ε(0,ε) small, considering each sequence {uin}H1(R3) as stated in Proposition 4, we have ui0 and Jε(ui)v=0 for all vC0(R3), i.e. Jε has a nontrival weak solution ui. Moreover, for i{1,,l},

    Qε(uin)Qε(ui). (4.5)

    And since

    Qε(uin)¯Bρ0(zi),nN,

    we have

    Qε(ui)¯Bρ0(zi). (4.6)

    Proof. By Proposition 4, ui0, i{1,,l} and uinui in Lp1oc(R3) for p[2,2), we obtain that

    uinlog|uin|2vdxuilog|ui|2vdx,vC0(R3)

    Besides, as in Proposition 4 and (4.4), we have

    limnR3(Iα|uin|p)|uin|p1v(Iα|uinui|p)|uinui|p1v=R3(Iα|ui|p)|ui|p1v,

    for all vC0(R3). And since

    (uinv+(V(εx)+1)uinv)dx(uiv+V(εx)+1)uiv)dx,

    for all vC0(R3). We conclude that Jε(ui)v=0 for all vC0(R3). By definition of g we have g(x)0 as |x|+, it is clear that

    χ(εx)g(εx)|uin|2dxχ(εx)g(εx)|ui|2dx

    and

    g(εx)|uin|2dxg(εx)|ui|2dx.

    Under the condition that these two limits hold, (4.5) and (4.6) are guaranteed.

    Next, we give a proof of Theorem 1, that is, there exist l solutions uiH1(R3){0}.

    Proof of Theorem 1.

    According to Corollary 4, for i{1,,l} and ε(0,ε), there exists a solution uiH1(R3){0} for problem (1.2) such that

    Qε(ui)¯Bρ0(zi).

    Because we have

    ¯Bρ0(zi)¯Bρ0(zj)=ϕ,ij.

    Then it implies that uiuj for ij.

    The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.

    The author declares there is no conflict of interest.



    [1] T. Cazenave, Stable solutions of the logarithmic Schrödinger equation, Nonlinear Anal., 7 (1983), 1127–1140. https://doi.org/10.1016/0362-546X(83)90022-6 doi: 10.1016/0362-546X(83)90022-6
    [2] T. Cazenave, An introduction to nonlinear Schrödinger equations, Universidade federal do Rio de Janeiro, Centro de ciências matemáticas e da natureza, Instituto de matemática, 1989.
    [3] T. Cazenave, A. Haraux, Équations d'évolution avec non linéarité logarithmique, Annales de la Faculté des sciences de Toulouse: Mathématiques, 2 (1980), 21–51. https://doi.org/10.5802/afst.543 doi: 10.5802/afst.543
    [4] I. Bialynicki-Birula, J. Mycielski, Nonlinear wave mechanics, Ann Phys-New York, 100 (1976), 62–93. https://doi.org/10.1016/0003-4916(76)90057-9
    [5] I. Bialynicki-Birula, J. Mycielski, Gaussons: solitons of the logarithmic Schrödinger equation, Phys Scripta, 20 (1979), 539. https://doi.org/10.1088/0031-8949/20/3-4/033 doi: 10.1088/0031-8949/20/3-4/033
    [6] I. Bialynicki-Birula, J. Mycielski, Wave equations with logarithmic nonlinearities, Bull. Acad. Pol. Sci. Cl., 3 (1975), 461–466.
    [7] W. Lian, M. S. Ahmed, R. Xu, Global existence and blow up of solution for semilinear hyperbolic equation with logarithmic nonlinearity, Nonlinear Anal., 184 (2019), 239–257. https://doi.org/10.1016/j.na.2019.02.015 doi: 10.1016/j.na.2019.02.015
    [8] X. Wang, Y. Chen, Y. Yang, J. Li, R. Xu, Kirchhoff-type system with linear weak damping and logarithmic nonlinearities, Nonlinear Anal., 188 (2019), 475–499. https://doi.org/10.1016/j.na.2019.06.019 doi: 10.1016/j.na.2019.06.019
    [9] Y. Chen, R. Xu, Global well-posedness of solutions for fourth order dispersive wave equation with nonlinear weak damping, linear strong damping and logarithmic nonlinearity, Nonlinear Anal., 192 (2020), 111664. https://doi.org/10.1016/j.na.2019.111664 doi: 10.1016/j.na.2019.111664
    [10] R. Carles, C. Su, Numerical study of the logarithmic Schrödinger equation with repulsive harmonic potential, Discrete Contin. Dyn. Syst. Ser. B, 28 (2023), 3136–3159. https://doi.org/10.3934/dcdsb.2022206 doi: 10.3934/dcdsb.2022206
    [11] E. H. Lieb, M. Loss, Analysis, volume 14. American Mathematical Soc., 2001. https://doi.org/10.1090/gsm/014
    [12] C. O. Alves, D. C. de Morais Filho, Existence and concentration of positive solutions for a Schrödinger logarithmic equation, Z. Angew. Math. Phys., 69 (2018), 1–22. https://doi.org/10.1007/s00033-018-1038-2 doi: 10.1007/s00033-018-1038-2
    [13] C. O. Alves, C. Ji, Multiple positive solutions for a Schrödinger logarithmic equation, Discrete Cont Dyn-A, preprint, arXiv: 1901.10329.
    [14] L. Guo, Y. Sun, G. Shi, Ground states for fractional nonlocal equations with logarithmic nonlinearity, Opusc. Math., 42 (2022), 157–178. https://doi.org/10.7494/OpMath.2022.42.2.157 doi: 10.7494/OpMath.2022.42.2.157
    [15] I. Bachar, H. Mli, H. Eltayeb, Nonnegative solutions for a class of semipositone nonlinear elliptic equations in bounded domains of Rn, Opusc. Math., 42, 2022. https://doi.org/10.7494/OpMath.2022.42.6.793 doi: 10.7494/OpMath.2022.42.6.793
    [16] A. Szulkin, Minimax principles for lower semicontinuous functions and applications to nonlinear boundary value problems, Ann I H Poincare-An, 3 (1986), 77–109. https://doi.org/10.1016/S0294-1449(16)30389-4 doi: 10.1016/S0294-1449(16)30389-4
    [17] B. Ilan, G. Fibich, G. Papanicolaou, Self-focusing with fourth-order dispersion, SIAM. J. Appl. Math., 62 (2002), 1437–1462. https://doi.org/10.1137/S0036139901387241 doi: 10.1137/S0036139901387241
    [18] B. Pausader, S. Shao, The mass-critical fourth-order Schrödinger equation in high dimensions, J Hyperbolic Differ Eq, 7 (2010), 651–705. https://doi.org/10.1142/S0219891610002256 doi: 10.1142/S0219891610002256
    [19] H. Brézis, E. Lieb, A relation between pointwise convergence of functions and convergence of functionals, P. Am. Math. Soc., 88 (1983), 486–490. https://doi.org/10.1090/S0002-9939-1983-0699419-3 doi: 10.1090/S0002-9939-1983-0699419-3
    [20] C. Ji, A. Szulkin, A logarithmic Schrödinger equation with asymptotic conditions on the potential, J. Math. Anal. Appl., 437 (2016), 241–254. https://doi.org/10.1016/j.jmaa.2015.11.071 doi: 10.1016/j.jmaa.2015.11.071
    [21] M. Squassina, A. Szulkin, Multiple solutions to logarithmic Schrödinger equations with periodic potential, Calc. Var. Partial Dif., 54 (2015), 585–597. https://doi.org/10.1007/s00526-014-0796-8 doi: 10.1007/s00526-014-0796-8
    [22] M. Willem, Minimax theorems, volume 24, Springer Science & Business Media, 1997. https://doi.org/10.1007/978-1-4612-4146-1
    [23] D. Cao, E. S. Noussair, Multiplicity of positive and nodal solutions for nonlinear elliptic problems in rn, Ann I H Poincare-An, 13 (1996), 567–588. https://doi.org/10.1016/S0294-1449(16)30115-9 doi: 10.1016/S0294-1449(16)30115-9
    [24] C. Mercuri, V. Moroz, J Van Schaftingen, Groundstates and radial solutions to nonlinear Schrödinger–poisson–slater equations at the critical frequency, Calc. Var. Partial Dif., 55 (2016), 1–58. https://doi.org/10.1007/s00526-016-1079-3 doi: 10.1007/s00526-016-1079-3
  • Reader Comments
  • © 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(999) PDF downloads(79) Cited by(0)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog