Home Standing waves for Choquard equation with noncritical rotation
Article Open Access

Standing waves for Choquard equation with noncritical rotation

  • Yicen Mao , Jie Yang and Yu Su EMAIL logo
Published/Copyright: June 5, 2024

Abstract

We investigate the existence and stability of standing waves with prescribed mass c > 0 for Choquard equation with noncritical rotation in Bose-Einstein condensation. Then, we consider the mass collapse behavior of standing waves, the ratio of energy to mass and the Lagrange multiplier, as c 0 + . Our results extend the existing results.

MSC 2010: 47G20; 35R11; 35A15; 46E30; 46E35

1 Introduction

In last two decades, the impressive progress of cold atom physics [42] gave a new impetus to the theory of many-bosons systems: the study of Bose-Einstein condensate, we refer to [18]. In the mean-field approximation, Lewin et al. [26,27] described a Bose-Einstein condensate of nonrelativistic particles with rotation of the atoms by the following equation:

i ψ t 1 2 ( i A ) 2 ψ + V ( x ) ψ = ( W * ψ p ) ψ p 2 ψ , ( t , x ) R + × R 3 , ψ ( 0 , x ) = ψ 0 ( x ) , ( C M )

where p = 2 and W is a uniformly bounded interaction potential. The potential V ( x ) has to incorporate a contribution due to the centrifugal force. Thus, one may take

(1.1) V ( x ) x 2 2 Ω 2 4 ( x 1 2 + x 2 2 ) ,

and Ω 0 is the rotational speed. The potential A : R 3 R 3 can model the Coriolis force due to the rotation of the atoms, where

A = Ω 2 ( x 2 , x 1 , 0 ) , x = ( x 1 , x 2 , x 3 ) R 3 , Ω 0 .

Then, we have the following useful identity:

(1.2) ( i A ) 2 = Δ 2 i A + A 2 i div A = Δ i Ω ( x 1 x 2 x 2 x 1 ) + Ω 2 4 ( x 1 2 + x 2 2 ) .

Putting equations (1.1) and (1.2) into ( C M ), one has

i ψ t 1 2 Δ ψ + 1 2 x 2 ψ L Ω ψ = ( W * ψ p ) ψ p 2 ψ , ( t , x ) R + × R 3 , ψ ( 0 , x ) = ψ 0 ( x ) . ( C B )

The rotation operator L Ω is

L Ω i Ω ( x 1 x 2 x 2 x 1 ) .

If Ω = 0 , then equation ( C B ) goes back to the description of the quantum theory of a polaron at rest by Pekar [40], and by Choquard on the modeling of an electron trapped in its own hole, as a certain approximation to the Hartree-Fock theory of one-component plasma [41]. For more physical interpretations with Ω = 0 , we refer to the study by Diósi [16].

If the interaction potential W is different, then equation ( C B ) will describe a different phenomenon. If one uses the contact interaction W = a δ 0 , which is proportional to the delta function at the origin, then equation ( C B ) goes back to the Gross-Pitaevskii equation with an angular momentum rotation as follows

i ψ t 1 2 Δ ψ + 1 2 x 2 ψ L Ω ψ = a ψ 2 p 2 ψ , ( t , x ) R + × R 3 , ψ ( 0 , x ) = ψ 0 ( x ) . ( C δ )

An appealing aspect of cold atoms experiments is the possibility to tune the value and even the sign of a , going from repulsive ( a > 0 ) to attractive interactions ( a < 0 ). In the mean-field regime, equation ( C δ ) is a Gross-Pitaevskii limit of a many-body Hamiltonian by Lieb et al. [31]. Moreover, Lieb et al. [32] proved a rigorous derivation of the Gross-Pitaevskii energy functional for two-dimensional case. For more physical interpretations, we refer to the study by Seiringer [44].

If one uses W = a δ 0 + W dip , where W dip = μ 0 μ dip 2 4 π 1 3 cos 2 θ x 3 is the long-range dipolar interaction potential between two dipoles ( μ 0 is the vacuum magnetic permeability, μ dip is the permanent magnetic dipole moment, and θ is the angle between the dipole axis and the vector x ), then equation ( C B ) with p = 2 becomes

i ψ t 1 2 Δ ψ + 1 2 x 2 ψ L Ω ψ = a ψ 2 ψ + ( W dip * ψ 2 ) ψ , ( t , x ) R + × R 3 , ψ ( 0 , x ) = ψ 0 ( x ) .

Equation ( C B ) arises in dipolar Bose-Einstein condensate in the unstable regime [48]. For more physical interpretations of Ω = 0 , we refer to [5,8,43].

As mentioned above, if the interaction potential W is different, then equation ( C B ) will describe a different phenomenon. In this article, we study equation ( C B ) with the following potential:

  1. The interaction potential W = I α is the Riesz potential as follows:

    I α ( x ) Γ 3 α 2 2 α π 3 2 Γ 3 2 x 3 α , x R 3 \ { 0 } ,

    where α ( 0 , 3 ) .

Under assumption ( P 1 ), equation ( C B ) becomes

i ψ t 1 2 Δ ψ + 1 2 x 2 ψ L Ω ψ = ( I α * ψ p ) ψ p 2 ψ , ( t , x ) R + × R 3 , ψ ( 0 , x ) = ψ 0 ( x ) . ( C )

In this article, we consider the existence, stability, and mass collapse behavior results for equation ( C ). Before explaining our contribution and the challenges encountered, let us first introduce some notations. We denote the norm of L p ( R 3 , R ) for any 1 p < as

u L p ( R 3 , R ) p = u L p ( R 3 , C ) p R 3 u p d x .

The basic Sobolev space is

D 1 , 2 ( R 3 , C ) u L 2 * ( R 3 , C ) u D 1 , 2 ( R 3 , C ) 2 R 3 u 2 d x < , 2 * = 6 .

Our working space is defined as follows:

W 2 1 , 2 ( R 3 , C ) u D 1 , 2 ( R 3 , C ) R 3 x 2 u 2 d x < ,

which is a Hilbert space with the inner product and norm

u , v W 2 1 , 2 ( R 3 , C ) Re R 3 u v ¯ d x + R 3 x 2 u v ¯ d x , u W 2 1 , 2 ( R 3 , C ) 2 u D 1 , 2 ( R 3 , C ) 2 + R 3 x 2 u 2 d x ,

where “ Re ” is the real part and v ¯ is the conjugate of v . We also need the following space:

Σ u W 2 1 , 2 ( R 3 , C ) R 3 u 2 d x < ,

with the norm

u Σ 2 u W 2 1 , 2 ( R 3 , C ) 2 + R 3 u 2 d x .

Clearly, the space Σ is equivalent to the space W 2 1 , 2 ( R 3 , C ) by Lemma 2.3.

An important topic on equation ( C ) is to study their standing wave solutions, which is a solution to the form ψ ( t , x ) = e i ω t u ( x ) for some ( u , ω ) W 2 1 , 2 ( R 3 , C ) × R such that u weakly solves

(1.3) 1 2 Δ u + 1 2 x 2 u L Ω u = ( I α * u p ) u p 2 u + ω u , x R 3 .

There are two different approaches to study the existence of solutions to equation (1.3) using variational methods:

  1. the frequency ω is a fixed and assigned parameter;

  2. the frequency ω is unknown.

For case (i), the solutions to equation (1.3) can be obtained as the critical points of the following energy functional defined in W 2 1 , 2 ( R 3 , C ) by

I ω ( u ) = 1 4 u W 2 1 , 2 ( R 3 , C ) 2 1 2 R 3 u ¯ L Ω u d x 1 2 p R 3 ( I α * u p ) u p d x ω 2 R 3 u 2 d x .

This case has attracted much attention in the last years; the existence and regularity of solutions with subcritical exponent was explored by Moroz and Van Schaftingen [38] and Cingolani and Tanaka [9], with critical exponent was studied by Gao and Yang [19], Su and Liu [47], and Giacomoni et al. [20], general nonlinearity was presented by Moroz and Van Schaftingen [39], and with double critical exponents was considered by Li and Ma [29]. For more results, we refer to [17,30,50].

For case (ii), the frequency ω is unknown, and one may consider the standing wave solution with prescribed mass c > 0 . Physically, this type of solution is a normalized solution. Normalized solutions to equation (1.3) can be obtained by searching critical points of the following energy functional:

J ( u ) = 1 4 u W 2 1 , 2 ( R 3 , C ) 2 1 2 R 3 u ¯ L Ω u d x 1 2 p R 3 ( I α * u p ) u p d x ,

on the constraint

S c { u W 2 1 , 2 ( R 3 , C ) u L 2 ( R 3 , R ) 2 = c } ,

with Lagrange multiplier ω . It is easy to see that J is well defined and C 1 on S c for p ( 3 + α 3 , 3 + α ) . We point out that 3 + α 3 is the energy lower critical exponent, and 3 + α is the energy upper critical exponent. The following mass critical exponent plays a key role in this article:

p ¯ = 5 + α 3 .

If Ω = 0 and p = α = 2 , Lieb [33] studied the existence and uniqueness of normalized solutions by using symmetrization techniques, and Lions [35,36] showed the existence and stability results. The existence of solutions with mass subcritical exponent was condidered by Cingolani and Tanaka [11] and Cingolani et al. [10], with lower Sobolev critical exponent was studied by Yao et al. [51]. If Ω = 0 and p ( 5 + α 3 , 3 + α ) , Li and Ye [28] investigated the existence of mountain pass-type normalized solution.

If u C 0 ( R 3 ) and α 3 , then equation (1.3) becomes (since I α * u p u p , for all u C 0 ( R 3 ) )

(1.4) 1 2 Δ u + 1 2 x 2 u L Ω u = u 2 p 2 u + ω u , x R 3 .

Normalized solutions to equation (1.4) can be obtained by searching critical points of the following energy functional (set q = 2 p ):

H ( u ) = 1 4 R 3 u 2 d x + 1 4 R 3 x 2 u 2 d x 1 2 R 3 u ¯ L Ω u d x 1 q R 3 u q d x ,

on S c with Lagrange multiplier ω . The corresponding mass critical exponent is

q ¯ = 10 3 .

For Ω = 0 , the existence of normalized solution to equation (1.4) with energy critical was explored by Jeanjean and Le [25]. Moreover, Jeanjean et al. [24] showed the stability results. For more works, we refer to [12,46]. For Ω > 0 , there are two cases:

  1. Ω ( 0 , 1 ) is the noncritical rotational speed;

  2. Ω = 1 is the critical rotational speed.

Arbunich et al. [2] considered the existence, stability, and instability of the standing waves with noncritical rotational speed. Bellazzini et al. [4] established a new local minimization method for equation ( C δ ) with a partial harmonic confinement and obtained the existence and stability results for mass-supercritical. Then, inspired by the method of Bellazzini et al. [4], Ardila and Hajaiej [3] and Luo and Yang [37] considered the case of noncirtical rotational speed and mass-supercritical. Cao et al. [7] investigated the X-ray free electron laser Schrödinger equation. Dinh [13] established the existence and stability results for critical rotational speed. For more results, we refer to [14,15, 21,22].

To the best of our knowledge, there is no result concerning the existence, stability, and mass collapse results for equation ( C ) with noncritical rotational speed Ω ( 0 , 1 ) . Hence, we focus our attention on the existence, stability, and mass collapse results for equation ( C ).

1.1 Main results

We give the definition of a ground-state in the following sense.

Definition 1.1

We say that u is a normalized ground-state solution of equation ( C ) if it satisfies

( J S c ) ( u ) = 0 and J ( u ) = inf { J ( v ) v S c and ( J S c ) ( v ) = 0 } .

First of all, we consider the mass subcritical case p ( 3 + α 3 , 5 + α 3 ) . The functional J is bounded from below on S c , and the minimization problem

m c inf v S c J ( v )

is achieved.

Theorem 1.1

Let α ( 0 , 3 ) , 3 + α 3 < p < 5 + α 3 , and 0 < Ω < 1 . Then, for any c > 0 , we have the following results.

  1. There exists a couple of weak solution ( u c , ω c ) S c × R to equation ( C ). Moreover, u c is a normalized ground-state solution.

  2. We have ω c < 3 .

  3. We know u c 0 in W 2 1 , 2 ( R 3 , C ) , as c 0 + , and

    sup u N c u l 0 φ 0 Σ 2 = O ( c + c p ) ,

    where φ 0 is the eigenfunction of the harmonic oscillator Δ + x 2 and l 0 = R 3 u φ 0 d x .

  4. There exists ω ¯ [ 2 m 1 , 3 ] such that

    ω ¯ 2 = lim c 0 + m c c = lim c 0 + ω c 2 = lim c 0 + 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x 2 c = lim c 0 + R 3 u c 2 d x R 3 u ¯ c L Ω u c d x 2 c = lim c 0 + R 3 x 2 u c 2 d x R 3 u ¯ c L Ω u c d x 2 c .

    where m 1 inf v S 1 J ( v ) .

Remark 1.1

Our method can be extended to Schrödinger equation (1.4). The results of the mass collapse behavior of standing waves lim c 0 + u c , the ratio of energy to mass lim c 0 + m c c , and the Lagrange multiplier lim c 0 + ω c are also new for it.

For p ( 5 + α 3 , 3 + α ) , we define the set of all ground-state with a given mass c as follows:

N c { u S c J ( u ) = m c } .

It follows from Theorem 1.1 that N c . We then study the stability of N c for 3 + α 3 < p < 5 + α 3 .

Theorem 1.2

Let α ( 0 , 3 ) , 3 + α 3 < p < 5 + α 3 , and 0 < Ω < 1 . Then, N c is stable in Σ .

Note that m c = for p ( 5 + α 3 , 3 + α ) . Clearly, for the case p ( 5 + α 3 , 3 + α ) , the functional J is unbounded from below on S c . Then, the global minimization method in Theorem 1.1 does not work. Motivated by [4], and recent works [3,7, 13,37], we study a local minimization problem:

m c , r inf v S c B r J ( v ) ,

where

B r { u W 2 1 , 2 ( R 3 , C ) u W 2 1 , 2 ( R 3 , C ) 2 r } .

For any fixed r > 0 , there exists c 0 c 0 ( r , Ω , p , α ) > 0 such that

( C 1 ) S c B r c < c 0 , ( C 2 ) c , r B r c < c 0 ,

where c , r { u S c B r J ( u ) = m c , r } . The claim ( C 1 ) guarantees that m c , r > , and m c , r is achieved. The claim ( C 2 ) is crucial to show that the minimizers of m c , r do not belong to the boundary of S c B r . Then, one can make sure that the minimizers of m c , r are critical points of the functional J S c .

Theorem 1.3

Let α ( 0 , 3 ) , 5 + α 3 p < 3 + α , and 0 < Ω < 1 . For any fixed r > 0 , we can find

0 < c 0 c 0 ( r , Ω , p , α ) = min Ω + 1 6 r , Ω 2 ( 1 Ω 2 ) 24 , 1 Ω 2 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 2 3 + α p , × p ( 1 Ω ) ( 1 Ω 2 ) 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 2 3 + α p ,

where C α is the constant in Lemma 2.5, C 6 p 3 + α is the constant in Lemma 2.1. Then, we have the following results.

  1. For any c ( 0 , c 0 ) , there exists a couple of weak solution ( u c , ω c ) c , r × R to equation ( C ).

  2. The lower and upper bounds of ω c are as follows:

    3 1 Ω 2 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 c 3 + α p 2 ω c < 3 .

  3. Moreover, we know

    sup u c , r u l 0 φ 0 Σ 2 = O ( c + c 3 + α p 2 ) ,

    where φ 0 is the eigenfunction of the harmonic oscillator Δ + x 2 and l 0 = R 3 u φ 0 d x .

In Theorem 1.3, we do not know whether u c is a normalized ground-state solution or not. Next, we show that u c is also a normalized ground-state solution if c > 0 is sufficiently small. Moreover, we also consider the mass collapse behavior of standing waves, the ratio of energy to mass, and the Lagrange multiplier, as c 0 + .

Theorem 1.4

Let α ( 0 , 3 ) , p ( 5 + α 3 , 3 + α ) , and 0 < Ω < 1 . Let ( u c , ω c ) c , r × R be given in Theorem 1.3. Then, we have the following results:

  1. We know that u c is a normalized ground-state solution for c > 0 sufficiently small. Moreover, u c 0 in W 2 1 , 2 ( R 3 , C ) , as c 0 + .

  2. There exists ω ¯ ¯ 3 ( 1 Ω 2 ) 8 , 3 such that

    ω ¯ ¯ 2 = lim c 0 + m c , r c = lim c 0 + ω c 2 = lim c 0 + 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x 2 c = lim c 0 + R 3 u c 2 d x R 3 u ¯ c L Ω u c d x 2 c = lim c 0 + R 3 x 2 u c 2 d x R 3 u ¯ c L Ω u c d x 2 c .

Remark 1.2

Our method can be extended to Schrödinger equation (1.2), we omit it. Compared with Luo and Yang [37], they restricted the parameter 0 < Ω < 1 ( 4 3 ( p 2 ) ) 2 . Our method can extended to their result to 0 < Ω < 1 .

We also study the stability of c , r for 5 + α 3 < p < 3 + α , and c > 0 small.

Theorem 1.5

Let α ( 0 , 3 ) , 5 + α 3 < p < 3 + α , 0 < Ω < 1 , and c > 0 small. Then, c , r is stable in Σ .

For 5 + α 3 < p < 3 + α , we show that there is a mountain pass-type solution.

Theorem 1.6

Let α ( 0 , 3 ) , 5 + α 3 < p < 3 + α , and 0 < Ω < 1 . Let c < c 0 be given in Theorem 1.3. Then, there exist ( u ¯ c , ω ¯ c ) Σ × R such that u ¯ c is the weak solution to equation ( C ) with ω ¯ c and J ( u ¯ c ) > m c , r .

Remark 1.3

For the existence of mountain pass-type solution, our method can be extended to Schrödinger equation (1.2), we omit it. Compared with Luo and Yang [37], they restricted the parameter 0 < Ω < 1 ( 4 3 ( p 2 ) ) 2 . Our method can extend to their result to 0 < Ω < 1 .

The article is organized as follows: in Section 2, we present some preliminaries; in Section 3, we prove the Pohozaev-type identity; in Section 4, we give the proof of Theorems 1.1 and 1.2; in Section 5, we prove Theorem 1.3; in Section 6, we present the proof of Theorems 1.4 and 1.5; in Section 7, we show Theorem 1.6.

2 Preliminaries

Lemma 2.1

[45] The following continuous and compactness embeddings hold

W 2 1 , 2 ( R 3 , C ) L s ( R 3 ) , s [ 2 , 2 * ] , W 2 1 , 2 ( R 3 , C ) L s ( R 3 ) , s [ 2 , 2 * ) .

Lemma 2.2

[49, Gagliardo-Nirenberg inequality] Let p ( 2 , 2 * ) . Then, there exists a constant C p > 0 such that

u L p ( R 3 , R ) C p u D 1 , 2 ( R 3 , R ) 3 ( p 2 ) 2 p u L 2 ( R 3 , R ) 1 3 ( p 2 ) 2 p u H 1 ( R 3 , R ) ,

where C p = p 2 W p L 2 ( R 3 ) p 2 1 p , W p is the ground-state solution of

Δ W + 2 p 3 ( p 2 ) 1 W = 4 3 ( p 2 ) W p 2 W .

Lemma 2.3

[49] Let x u L 2 ( R 3 , C ) and u D 1 , 2 ( R 3 , C ) . Then, u L 2 ( R 3 , R ) and

u L 2 ( R 3 , R ) 2 2 3 u D 1 , 2 ( R 3 , R ) x u L 2 ( R 3 , R ) ,

with equality holding for the functions u ( x ) = e 1 2 x 2 .

Note that u u for all u H 1 ( R 3 , C ) and u W 2 1 , 2 ( R 3 , C ) , see [34, Theorem 6.17]. Thus, Lemmas 2.2 and 2.3 are also true for u H 1 ( R 3 , C ) and u W 2 1 , 2 ( R 3 , C ) , respectively.

We show the following result, which plays a key role in this article.

Lemma 2.4

If 0 < Ω < 1 , then

1 Ω 2 2 u W 2 1 , 2 ( R 3 , C ) 2 u W 2 1 , 2 ( R 3 , C ) 2 2 R 3 u ( x ) ¯ L Ω u ( x ) d x 2 u W 2 1 , 2 ( R 3 , C ) 2 .

Proof

It follows from Hölder’s, Cauchy-Schwarz’s, and Young’s inequalities that

(2.1) R 3 u ( x ) ¯ L Ω u ( x ) d x = R 3 u ( x ) ¯ [ i Ω ( x 1 x 2 x 2 x 1 ) ( u ) ( x ) ] d x Ω R 3 x 1 2 u 2 d x 1 2 R 3 x 2 u 2 d x 1 2 + R 3 x 2 2 u 2 d x 1 2 R 3 x 1 u 2 d x 1 2 Ω R 3 ( x 1 2 + x 2 2 ) u 2 d x 1 2 R 3 ( x 1 u 2 + x 2 u 2 ) d x 1 2 ε R 3 ( x 1 u 2 + x 2 u 2 ) d x + Ω 2 4 ε R 3 ( x 1 2 + x 2 2 ) u 2 d x ε R 3 u 2 d x + Ω 2 4 ε R 3 x 2 u 2 d x ,

for any ε > 0 .

Applying equation (2.1) with ε = 1 2 , we have

u W 2 1 , 2 ( R 3 , C ) 2 2 R 3 u ( x ) ¯ L Ω u ( x ) d x u W 2 1 , 2 ( R 3 , C ) 2 + 2 R 3 u ( x ) ¯ L Ω u ( x ) d x u W 2 1 , 2 ( R 3 , C ) 2 + R 3 u 2 d x + Ω 2 2 R 3 x 2 u 2 d x 2 u W 2 1 , 2 ( R 3 , C ) 2 .

To prove the reverse inequality, we use equation (2.1) again

u W 2 1 , 2 ( R 3 , C ) 2 2 R 3 u ( x ) ¯ L Ω u ( x ) d x u W 2 1 , 2 ( R 3 , C ) 2 2 R 3 u ( x ) ¯ L Ω u ( x ) d x ( 1 2 ε ) R 3 u 2 d x + 1 Ω 2 2 ε R 3 x 2 u 2 d x .

We choose ε = 2 Ω 2 1 + Ω 2 > 0 . Then,

1 2 ε = 1 Ω 2 1 + Ω 2 and 1 Ω 2 2 ε = 1 Ω 2 2 .

It follows from 1 Ω 2 1 + Ω 2 > 1 Ω 2 2 that

u W 2 1 , 2 ( R 3 , C ) 2 R 3 u ( x ) ¯ L Ω u ( x ) d x 1 Ω 2 1 + Ω 2 R 3 u 2 d x + 7 Ω 2 8 R 3 x 2 u 2 d x 1 Ω 2 2 u W 2 1 , 2 ( R 3 , C ) 2 .

The proof is complete.□

We recall the Hardy-Littlewood-Sobolev inequality as follows:

Lemma 2.5

[34] Let s , t > 1 and α ( 0 , 3 ) with 1 s + 1 t = 1 + α 3 . Then, there exists C ( α , s , t ) > 0 such that for any u L s ( R 3 ) and v L t ( R 3 ) , there holds

R 3 R 3 u ( x ) v ( y ) x y N α d x d y C ( α , s , t ) u L s ( R 3 ) v L t ( R 3 ) .

If s = t , then we set C α C ( α , s , s ) .

3 Pohozaev-type identity

In this section, we establish the following Pohozaev-type identity:

Lemma 3.1

Let 3 + α 3 < p < 3 + α and 0 < Ω < 1 . If v W 2 1 , 2 ( R 3 , C ) is a solution of the following problem:

(3.1) 1 2 Δ v + 1 2 x 2 v L Ω v = ( I α * v p ) v p 2 v + ω v , x R 3 ,

Then, the Pohozaev-type identity

1 2 R 3 v 2 d x 1 2 R 3 x 2 v 2 d x + 3 + α 2 p 3 2 R 3 ( I α * v p ) v p d x = Re R 3 ( x v ¯ ) L Ω v d x + 3 2 R 3 v ¯ L Ω v d x .

As a consequence, it satisfies

P ( v ) 1 2 R 3 v 2 d x 1 2 R 3 x 2 v 2 d x + 3 + α 3 p 2 p R 3 ( I α * v p ) v p d x = 0 .

Proof

Multiplying equation (3.1) by x v ¯ and then integrate by parts and take real part, one has

(3.2) Re 1 2 R 3 ( x v ¯ ) Δ v d x + 1 2 R 3 ( x v ¯ ) x 2 v d x R 3 ( x v ¯ ) L Ω v d x = Re R 3 ( x v ¯ ) ( I α * v p ) v p 2 v d x + ω R 3 ( x v ¯ ) v d x .

It is easy to see that

(3.3) Re R 3 ( x v ¯ ) v d x = 3 2 R 3 v 2 d x ,

and

(3.4) Re R 3 ( x v ¯ ) Δ v d x = 1 2 R 3 v 2 d x ,

and

(3.5) Re R 3 ( x v ¯ ) x 2 v d x = 5 2 R 3 x 2 v 2 d x .

Calculating the first term on the right-hand side of equation (3.2), we know

Re R 3 ( x v ¯ ( x ) ) R 3 v ( y ) p x y 3 α d y v ( x ) p 2 v ( x ) d x = Re R 3 v ¯ ( x ) x R 3 v ( y ) p x y 3 α d y v ( x ) p 2 v ( x ) d x = 3 R 3 ( I α * v p ) v p d x + ( 3 α ) Re R 3 x ( x y ) R 3 v ( y ) p x y 5 α d y v ( x ) p d x + ( p 1 ) Re R 3 ( I α * v p ) x v ¯ ( x ) v ( x ) v ( x ) p 2 d x = 3 R 3 ( I α * v p ) v p d x + ( 3 α ) Re R 3 x ( x y ) R 3 v ( y ) p x y 5 α d y v ( x ) p d x + ( p 1 ) Re R 3 ( I α * v p ) x v ( x ) v ¯ ( x ) v ( x ) p 2 d x

which shows

Re R 3 ( x v ¯ ( x ) ) R 3 v ( y ) p x y 3 α d y v ( x ) p 2 v ( x ) d x = 3 p R 3 ( I α * v p ) v p d x + 3 α p Re R 3 x ( x y ) R 3 v ( y ) p x y 5 α d y v ( x ) p d x .

Similarly,

Re R 3 ( y v ¯ ( y ) ) R 3 v ( x ) p x y 3 α d x v ( y ) p 2 v ( y ) d y = 3 p R 3 ( I α * v p ) v p d y + 3 α p Re R 3 y ( y x ) R 3 v ( x ) p x y 5 α d x v ( y ) p d y

and consequently, we obtain

(3.6) Re R 3 ( x v ¯ ( x ) ) R 3 v ( y ) p x y 3 α d y v ( x ) p 2 v ( x ) d x = 3 + α p R 3 ( I α * v p ) v p d x .

Putting equations (3.3)–(3.6) into (3.2), one has

(3.7) 1 4 R 3 v 2 d x 5 4 R 3 x 2 v 2 d x Re R 3 ( x v ¯ ) L Ω v d x = 3 + α 2 p R 3 ( I α * v p ) v p d x 3 ω 2 R 3 v 2 d x .

Note that v is also a weakly solution of equation (3.1). We have

(3.8) 1 2 R 3 v 2 d x + 1 2 R 3 x 2 v 2 d x R 3 v ¯ L Ω v d x = R 3 ( I α * v p ) v p d x + ω R 3 v 2 d x .

Combining equation (3.7) and (3.8), we obtain

1 4 R 3 v 2 d x 5 4 R 3 x 2 v 2 d x + 3 + α 2 p R 3 ( I α * v p ) v p d x = 3 4 R 3 v 2 d x 3 4 R 3 x 2 v 2 d x + 3 2 R 3 v ¯ L Ω v d x + 3 2 R 3 ( I α * v p ) v p d x + Re R 3 ( x v ¯ ) L Ω v d x ,

which implies the Pohozaev-type identity

(3.9) 1 2 R 3 v 2 d x 1 2 R 3 x 2 v 2 d x + 3 + α 2 p 3 2 R 3 ( I α * v p ) v p d x = Re R 3 ( x v ¯ ) L Ω v d x + 3 2 R 3 v ¯ L Ω v d x .

It is easy to check that

R 3 ( x v ¯ ) L Ω v d x = R 3 ( x v ) L Ω v ¯ d x 3 R 3 v ¯ L Ω v d x

and

Re R 3 ( x v ¯ ) L Ω v d x = Re R 3 ( x v ) L Ω v ¯ d x

which induces

(3.10) Re R 3 ( x v ¯ ) L Ω v d x = 3 2 R 3 v ¯ L Ω v d x .

Inserting equation (3.10) into equation (3.9), we obtain the desired result.□

4 Proofs of Theorems 1.1 and 1.2

In this section, we prove Theorems 1.1 and 1.2, i.e., the existence, stability, and mass collapse results for mass subcritical case.

Lemma 4.1

Let 3 + α 3 < p < 5 + α 3 and 0 < Ω < 1 . Then, for any c > 0 , the functional J is bounded from below on S c , and m c > .

Proof

For any u S c , it follows from Lemmas 2.5 and 2.2 that

(4.1) R 3 ( I α * u p ) u p d x C α u L 6 p 3 + α ( R 3 , R ) 2 p C α C 6 p 3 + α 2 p u D 1 , 2 ( R 3 , R ) 3 p α 3 u L 2 ( R 3 , R ) 3 + α p = C α C 6 p 3 + α 2 p u D 1 , 2 ( R 3 , R ) 3 p α 3 c 3 + α p 2 C α C 6 p 3 + α 2 p u W 2 1 , 2 ( R 3 , C ) 3 p α 3 c 3 + α p 2 .

Applying equation (4.1) and Lemma 2.4, for any u S c , we obtain

(4.2) J ( u ) = 1 4 u W 2 1 , 2 ( R 3 , C ) 2 1 2 R 3 u ¯ L Ω u d x 1 2 p R 3 ( I α * u p ) u p d x 1 2 1 Ω 2 4 u W 2 1 , 2 ( R 3 , C ) 2 1 2 p R 3 ( I α * u p ) u p d x 1 2 1 Ω 2 4 u W 2 1 , 2 ( R 3 , C ) 2 1 2 p C α C 6 p 3 + α 2 p u W 2 1 , 2 ( R 3 , C ) 3 p α 3 c 3 + α p 2 .

Note that p ( 3 + α 3 , 5 + α 3 ) . Then, for any c > 0 , the functional J is bounded from below on S c , and m c > .□

Lemma 4.2

Let 3 + α 3 < p < 5 + α 3 and 0 < Ω < 1 . Let { u n } S c be the minimizing sequence of m c . Then, { u n } is a bounded sequence in W 2 1 , 2 ( R 3 , C ) .

Proof

From equation (4.2), we know that J is coercive on S c . This shows that any minimizing sequence of m c is a bounded sequence in W 2 1 , 2 ( R 3 , C ) .□

We are now in a position to prove Theorem 1.1.

Proof of Theorem 1.1

We divide this proof into five steps.

Step 1. In this step, we show the existence of a normalized ground-state solution to equation ( C ) on S c . We choose a minimizing sequence { u n } S c . From Lemma 4.2, one has that { u n } is a bounded sequence in W 2 1 , 2 ( R 3 , C ) . By Banach-Alaoglu theorem, there exists a weakly convergent subsequence { u n k } { u n } (also defined by { u n } ) such that, for some u W 2 1 , 2 ( R 3 , C ) ,

u n u in W 2 1 , 2 ( R 3 , C ) , u n u in R 3 .

The compact embedding of Lemma 2.1 shows that u n u strongly in L q ( R 3 , C ) , q [ 2 , 2 * ) . Hence, we know u S c and

R 3 u 2 d x = R 3 u n 2 d x = c .

Applying Lemma 2.4, one has that J is weakly lower semi-continuous. Therefore, we obtain

J ( u ) lim n J ( u n ) = m c J ( u ) ,

which shows

J ( u ) = m c ,

and

u n u in W 2 1 , 2 ( R 3 , C ) .

Let u c S c be the minimizer of m c . Then, u c is a critical point of J S c . So, there exists a Lagrange multiplier ω c R such that ( u c , ω c ) solves equation ( C ) in weak sense, i.e.,

ω c R 3 u ¯ c ϕ d x = 1 2 R 3 u ¯ c ϕ d x + 1 2 R 3 x 2 u ¯ c ϕ d x R 3 u ¯ c L Ω ϕ d x R 3 ( I α * u ¯ c p ) u ¯ c p 2 u ¯ c ϕ d x

for all ϕ W 2 1 , 2 ( R 3 , C ) .

Step 2. We estimate the upper bound of the Lagrange multiplier ω c in this step. From Antonelli et al. [1], we know that the pure point spectrum of the harmonic oscillator on R 3 is

(4.3) σ p ( Δ + x 2 ) = { λ k = 3 + 2 k k N } ,

and the corresponding eigenfunctions are given by Hermite functions (the eigenvalue is λ k , and the eigenfunctions is φ k ), which are the orthonormal basis of L 2 ( R 3 , R ) . As φ 0 is real-valued, we know that φ 0 W 2 1 , 2 ( R 3 , C ) . Note that λ 0 = 3 and φ 0 S 1 . Then, φ c φ 0 S c . It follows from Lemma 2.4 and φ is real-valued that

(4.4) m c J ( φ ) = 1 4 φ W 2 1 , 2 ( R 3 , C ) 2 1 2 R 3 φ ¯ L Ω φ d x 1 2 p R 3 ( I α * φ p ) φ p d x 1 2 φ W 2 1 , 2 ( R 3 , C ) 2 1 2 p R 3 ( I α * φ p ) φ p d x < 1 2 φ W 2 1 , 2 ( R 3 , C ) 2 = 3 2 c .

Note that ( u c , ω c ) c × R weakly solves equation ( C ). From equation (4.4), we have

c ω c = ω c u c L 2 ( R 3 , R ) 2 = 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x R 3 ( I α * u c p ) u c p d x = 2 J ( u c ) + 1 p p R 3 ( I α * u c p ) u c p d x < 2 J ( u c ) = 2 m c < 3 c ,

which implies

ω c < 3 .

Step 3. We study the mass collapse behavior of m c c as c 0 + . Before that, we establish the following inequality, which plays a key role in this step:

(4.5) m c c < m c ¯ c ¯ ,

where 0 < c ¯ < c < , 3 + α 3 < p < 5 + α 3 , and 0 < Ω < 1 .

Let { u n } S c ¯ be a bounded minimizing sequence of m c ¯ . Then, for any c c ¯ > 1 , we have

c c ¯ R 3 u n 2 d x = c c c ¯ u n S c .

Then, we have

J c c ¯ u n c c ¯ J ( u n ) = 1 2 p c c ¯ c c ¯ p R 3 ( I α * u n p ) u n p d x .

From Step 1 of Theorem 1.1, we know that there exists C > 0 such that

lim n R 3 ( I α * u n p ) u n p d x C > 0 .

Note that c c ¯ > 1 and p > 3 + α 3 > 1 . Then, one has

J c c ¯ u n < c c ¯ J ( u n ) .

It follows from m c J c c ¯ u n and m c ¯ = J ( u n ) that

m c J c c ¯ u n < c c ¯ J ( u n ) = c c ¯ m c ¯ .

Then, we have the inequality (4.5).

It is easy to see that

2 m c + 1 p p R 3 ( I α * u c p ) u c p d x 2 c = 2 J ( u c ) + 1 p p R 3 ( I α * u c p ) u c p d x 2 u c L 2 ( R 3 , R ) 2 = 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x R 3 ( I α * u c p ) u c p d x 2 u c L 2 ( R 3 , R ) 2 = ω c 2 ,

which gives

(4.6) m c c + 1 p p R 3 ( I α * u c p ) u c p d x 2 c = ω c 2 .

Inequality (4.5) implies that m c c is monotone decreasing. Then,

< m 1 1 < m c c , for all c ( 0 , 1 ) .

It follows from equation (4.4) that

m c c < 3 2 .

Then, by using the monotone bounded convergence theorem, we know that the limitation lim c 0 + m c c exists. Moreover, we know

lim c 0 + m c c m 1 , 3 2 .

Then, there exist ω ¯ [ 2 m 1 , 3 ] such that

lim c 0 + m c c = ω ¯ 2 .

Step 4. In this step, we estimate the mass collapse behavior of u c as c 0 + . For any u N c , we write u = u 1 + i u 2 , where u 1 is the real part and u 2 is the imaginary part of u . Then we obtain

u = k = 0 R 3 u 1 φ k d x φ k + i k = 0 R 3 u 2 φ k d x φ k = k = 0 l k φ k ,

where φ k is the eigenfunctions of λ k in equation (4.3), and l k = R 3 u φ k d x . Thus,

(4.7) c = R 3 u 2 d x = k = 0 l k l ¯ k R 3 φ k 2 d x = k = 0 l k 2 ,

where l ¯ k is the conjugate of l k .

Note that u N c . Then,

1 2 R 3 u 2 d x + 1 2 R 3 x 2 u 2 d x R 3 u ¯ L Ω u d x R 3 ( I α * u p ) u p d x ω R 3 u 2 d x = 0 .

Thus,

J ( u ) = 1 2 1 2 1 2 p u W 2 1 , 2 ( R 3 , C ) 2 2 R 3 u ¯ L Ω u d x + ω 2 p R 3 u 2 d x 1 2 p 1 p 1 Ω 2 4 u W 2 1 , 2 ( R 3 , C ) 2 + ω 2 p R 3 u 2 d x ,

which gives, by using R 3 u 2 d x = c and 0 ω < 3 as c 0 + ,

(4.8) u W 2 1 , 2 ( R 3 , C ) 2 4 ( 1 Ω 2 ) ( p 1 ) 2 p J ( u ) ω R 3 u 2 d x = 4 ( 1 Ω 2 ) ( p 1 ) [ 2 p J ( u ) ω c ] .

We show that u c W 2 1 , 2 ( R 3 , C ) c is bounded, as c 0 + . Assume on the contrary that u c W 2 1 , 2 ( R 3 , C ) c . By using 3 2 > m c c , (4.2), and 2 > 3 p α 3 , as c 0 + , we have

3 2 > m c c = J ( u c ) c 1 Ω 2 8 u c W 2 1 , 2 ( R 3 , C ) 2 c C u c W 2 1 , 2 ( R 3 , C ) 3 p α 3 c 3 + α p 2 c = 1 Ω 2 8 u c W 2 1 , 2 ( R 3 , C ) c 2 C u c W 2 1 , 2 ( R 3 , C ) c 3 p α 3 c p 1 + .

This is a contradiction. Hence, u c W 2 1 , 2 ( R 3 , C ) c is bounded, as c 0 + , and there exists C > 3 independent of c such that

(4.9) u c W 2 1 , 2 ( R 3 , C ) C c .

Hence, u c 0 in W 2 1 , 2 ( R 3 , C ) , as c 0 + .

It follows from Lemma 2.4, (4.1), (4.3), and u W 2 1 , 2 ( R 3 , C ) 2 < 12 p ( 1 Ω 2 ) ( p 1 ) c that

3 2 c > J ( u ) = 1 4 u W 2 1 , 2 ( R 3 , C ) 2 1 2 R 3 u ¯ L Ω u d x 1 2 p R 3 ( I α * u p ) u p d x 1 Ω 2 8 u W 2 1 , 2 ( R 3 , C ) 2 1 2 p C α C 6 p 3 + α 2 p u W 2 1 , 2 ( R 3 , C ) 3 p α 3 c 3 + α p 2 1 Ω 2 8 k = 0 λ k l k 2 1 2 p C α C 6 p 3 + α 2 p C 3 p α 3 2 c p ,

which gives

3 k = 0 l k 2 = λ 0 k = 0 l k 2 k = 0 λ k l k 2 12 1 Ω 2 c + 4 p ( 1 Ω 2 ) C α C 6 p 3 + α 2 p C 3 p α 3 2 c p .

Thus, one has

u l 0 φ 0 W 2 1 , 2 ( R 3 , C ) 2 = k = 1 l k φ k W 2 1 , 2 ( R 3 , C ) 2 = k = 1 λ k l k 2 k = 0 λ k l k 2 12 1 Ω 2 c + 4 p ( 1 Ω 2 ) C α C 6 p 3 + α 2 p C 3 p α 3 2 c p

and

u l 0 φ 0 L 2 ( R 3 , R ) 2 = k = 1 l k φ k L 2 ( R 3 , R ) 2 = k = 1 l k 2 k = 0 l k 2 4 1 Ω 2 c + 4 3 p ( 1 Ω 2 ) C α C 6 p 3 + α 2 p C 3 p α 3 2 c p .

Combining the above two results, we have

sup u c , r u l 0 φ 0 Σ 2 4 4 1 Ω 2 c + 4 3 p ( 1 Ω 2 ) C α C 6 p 3 + α 2 p C 3 p α 3 2 c p .

Then, we obtain the desired result sup u N c u l 0 ψ 0 Σ 2 = O ( c + c p ) .

Step 5. In this step, we study the mass collapse behavior of ω c as c 0 + . And then from equations (4.1) and (4.9), one has

0 R 3 ( I α * u c p ) u c p d x c C u c W 2 1 , 2 ( R 3 , C ) 3 p α 3 c 3 + α p 2 c = C u c W 2 1 , 2 ( R 3 , C ) c 3 p α 3 c p 1 C c p 1 0 , as c 0 + .

By using the Sandwich theorem, we know that lim c 0 + R 3 ( I α * u c p ) u c p d x c = 0 . Putting this result with (4.6), we have

0 ω c 2 m c c 1 p p R 3 ( I α * u c p ) u c p d x 2 c 0 , as c 0 + .

This shows

lim c 0 + ω c 2 = lim c 0 + m c c = ω ¯ 2 .

Then

(4.10) ω ¯ = lim c 0 + ω c = lim c 0 + 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x R 3 ( I α * u c p ) u c p d x c = lim c 0 + 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x c .

It follows from Lemma 3.1 that

0 = P ( u c ) c = 1 2 R 3 u c 2 d x R 3 x 2 u c 2 d x c + 3 + α 3 p 2 p R 3 ( I α * u c p ) u c p d x c = 1 2 R 3 u c 2 d x R 3 x 2 u c 2 d x c , as c 0 + ,

which shows

(4.11) lim c 0 + R 3 u c 2 d x c = lim c 0 + R 3 x 2 u c 2 d x c .

Combining equations (4.10) and (4.11), we have

ω ¯ = lim c 0 + ω c = lim c 0 + 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x c = lim c 0 + R 3 u c 2 d x R 3 u ¯ c L Ω u c d x c = lim c 0 + R 3 x 2 u c 2 d x R 3 u ¯ c L Ω u c d x c .

We obtain the desired result.□

Lemma 4.3

Let 3 + α 3 < p < 5 + α 3 and 0 < Ω < 1 . Then, the corresponding solution to equation ( C ) exists globally in time.

Proof

Let u t ( x ) u ( t , x ) : ( T * , T * ) × R N C be the corresponding solution to equation ( C ). It follows from equation (4.1), p ( 3 + α 3 , 5 + α 3 ) , Young’s inequality, and the conservation of mass that

(4.12) 1 2 p R 3 ( I α * u t p ) u t p d x C u t W 2 1 , 2 ( R 3 , C ) 3 p α 3 u t L 2 ( R 3 , R ) 3 + α p 2 ε 3 p α 3 3 p α 3 2 u t W 2 1 , 2 ( R 3 , C ) 3 p α 3 2 3 p α 3 2 3 p α 3 + C 2 ε 3 p α 3 3 p α 3 2 u t L 2 ( R 3 , R ) 3 + α p 2 5 + α 3 p 2 5 + α 3 p ε u t W 2 1 , 2 ( R 3 , C ) 2 + C ˜ ε 3 + α p 5 + α 3 p u t L 2 ( R 3 , R ) 2 ( 3 + α p ) 5 + α 3 p = ε u t W 2 1 , 2 ( R 3 , C ) 2 + C ˜ ε 3 + α p 5 + α 3 p u 0 L 2 ( R 3 , R ) 2 ( 3 + α p ) 5 + α 3 p ,

for some constant C ˜ > 0 depending on α , p .

By Lemma 2.4, equation (4.12), and the energy conservation, for all t ( T * , T * ) , we obtain

J ( u 0 ) = J ( u t ) 1 Ω 2 8 u t W 2 1 , 2 ( R 3 , C ) 2 1 2 p R 3 ( I α * u t p ) u t p d x 1 Ω 2 8 ε u t W 2 1 , 2 ( R 3 , C ) 2 C ˜ ε 3 + α p 5 + α 3 p u 0 L 2 ( R 3 , R ) 2 ( 3 + α p ) 5 + α 3 p .

Taking ε = 1 Ω 2 16 , for all t ( T * , T * ) , we obtain

J ( u 0 ) 1 Ω 2 16 u t W 2 1 , 2 ( R 3 , C ) 2 C ˜ 1 Ω 2 16 3 + α p 5 + α 3 p u 0 L 2 ( R 3 , R ) 2 ( 3 + α p ) 5 + α 3 p ,

which gives

u t W 2 1 , 2 ( R 3 , C ) 2 C ˜ p , α , Ω , u 0 16 1 Ω 2 J ( u 0 ) + C ˜ 1 Ω 2 16 3 + α p 5 + α 3 p 1 u 0 L 2 ( R 3 , R ) 2 ( 3 + α p ) 5 + α 3 p .

From above, we know

sup t ( T * , T * ) u t W 2 1 , 2 ( R 3 , C ) 2 C ˜ p , α , Ω , u 0 .

Clearly, the blow-up cannot occur at finite time.□

Proof of Theorem 1.2

By way of contradiction, we assume that the set of ground-state N c is unstable. Then, there exists ε 0 > 0 , v 0 N c , and a sequence of initial data { v n 0 } Σ satisfying

v n 0 v 0 Σ 0 , as n ,

and a sequence of times { t n } R such that

inf v N c v n ( t n , ) v Σ > ε 0 ,

where v n ( 0 , ) = v n 0 ( ) . Note that the solutions exist globally in time by Lemma 4.3.

Without loss of generality, we may assume that { v n 0 } S c . For simplicity, set v ¯ n ( x ) v n ( t n , x ) . We have mass conservation, as n ,

R 3 v ¯ n 2 d x R 3 v n ( t n , x ) 2 d x = R 3 v n 0 ( x ) 2 d x R 3 v 0 ( x ) 2 d x = c

and energy conservation

J ( v ¯ n ) J ( v n ( t n , ) ) = J ( v n 0 ) J ( v 0 ) = m c .

Consequently, the continuity in time implies that v ¯ n is a minimizing sequence. By the proof of Theorem 1.1, there exists a subsequence such that v ¯ n k v ¯ Σ strongly. Thus,

inf v N c v n ( t n , ) v Σ v ¯ n k v ¯ Σ 0 , as n .

This is a contradiction.□

5 Proof of Theorem 1.3

In this section, we prove Theorem 1.3. We show that J S c presents a local minima structure, which guarantees that the minimizer of m c , r is a critical point of J S c .

Lemma 5.1

For any r > 0 , we obtain that S c B r if and only if c r 3 .

Proof

Let r > 0 be fixed. If S c B r , then for any u S c B r , by using Lemma 2.3 and Young’s inequality, we obtain

c = u L 2 ( R 3 , R ) 2 2 3 u D 1 , 2 ( R 3 , R ) x u L 2 ( R 3 , R ) 2 3 1 2 u D 1 , 2 ( R 3 , C ) 2 + 1 2 x u L 2 ( R 3 , C ) 2 1 3 u W 2 1 , 2 ( R 3 , C ) 2 r 3 .

On the other hand, let ϕ ( x ) = e 1 2 x 2 , and by using Lemma 2.3 again, we have

π 3 2 = ϕ L 2 ( R 3 , R ) 2 = 2 3 ϕ D 1 , 2 ( R 3 , R ) 2 = 2 3 x ϕ L 2 ( R 3 , R ) 2 .

Set ϕ c c π 3 4 ϕ = c π 3 4 e 1 2 x 2 . Then, one has

c = ϕ c L 2 ( R 3 , R ) 2 = ϕ c L 2 ( R 3 , C ) 2 ,

and

ϕ c D 1 , 2 ( R 3 , R ) 2 = x ϕ c L 2 ( R 3 , R ) 2 = 3 2 c .

For any c r 3 , we have

ϕ c D 1 , 2 ( R 3 , C ) 2 + x ϕ c L 2 ( R 3 , C ) 2 = ϕ c D 1 , 2 ( R 3 , R ) 2 + x ϕ c L 2 ( R 3 , R ) 2 = 3 c r ,

which gives ϕ c S c B r .□

Lemma 5.2

Let 5 + α 3 p < 3 + α and 0 < Ω < 1 . For any r > 0 , there exists

0 < c 0 c 0 ( r , Ω , p , α ) = min Ω + 1 6 r , Ω 2 ( 1 Ω 2 ) 24 , 1 Ω 2 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 2 3 + α p , p ( 1 Ω ) ( 1 Ω 2 ) 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 2 3 + α p

such that for any c < c 0 ,

inf u S c B ν r J ( u ) < inf u S c ( B r B μ r ) J ( u ) ,

where ν = Ω 2 ( 1 Ω 2 ) 8 and μ = Ω .

Proof

First, we show S c ( B r B μ r ) with c < c 0 < Ω + 1 6 r . From Lemma 5.1, we know that ϕ c S c B r and R 3 ϕ c 2 d x = R 3 x 2 ϕ c 2 d x = 3 2 c . Set ϕ c , ξ = ξ 3 2 ϕ c ( ξ x ) . Then,

ϕ c , ξ W 2 1 , 2 ( R 3 , C ) 2 = R 3 ϕ c , ξ 2 d x + R 3 x 2 ϕ c , ξ 2 d x = ξ 2 R 3 ϕ c 2 d x + ξ 2 R 3 x 2 ϕ c 2 d x = 3 2 c ( ξ 2 + ξ 2 ) .

Clearly, for any ξ > 0 , we have

ξ 2 + ξ 2 2 .

Note that c < Ω + 1 6 r . For each r > 0 , we can choose ξ 0 > 0 such that

2 < Ω + 1 3 c r = ( ξ 0 2 + ξ 0 2 ) .

Then, we have

ϕ c , ξ W 2 1 , 2 ( R 3 , C ) 2 = Ω + 1 2 r = μ + 1 2 r ( μ r , r ) .

This shows that for any r > 0 and ϕ c S c B r , there exists ϕ c , ξ 0 S c ( B r \ B μ r ) .

For any u S c ( B r \ B μ r ) , by using Lemma 2.4 and equation (4.1) that

J ( u ) = 1 4 R 3 u 2 d x + 1 4 R 3 x 2 u 2 d x 1 2 R 3 u ¯ L Ω u d x 1 2 p R 3 ( I α * u p ) u p d x 1 2 1 Ω 2 4 u W 2 1 , 2 ( R 3 , C ) 2 1 2 p R 3 ( I α * u p ) u p d x 1 2 1 Ω 2 4 u W 2 1 , 2 ( R 3 , C ) 2 1 2 p C α C 6 p 3 + α 2 p u W 2 1 , 2 ( R 3 , C ) 3 p α 3 c 3 + α p 2 = 1 2 u W 2 1 , 2 ( R 3 , C ) 2 1 Ω 2 4 1 p C α C 6 p 3 + α 2 p u W 2 1 , 2 ( R 3 , C ) 3 p α 5 c 3 + α p 2 1 2 μ r 1 Ω 2 4 1 p C α C 6 p 3 + α 2 p r 3 p α 5 2 c 3 + α p 2

where we restrict c < 1 Ω 2 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 2 3 + α p .

From Lemma 5.1, one has

S c B ν r if and only if c ν r 3 .

For any u S c B ν r , from Lemma 2.4, we obtain

J ( u ) = 1 4 R 3 u 2 d x + 1 4 R 3 x 2 u 2 d x 1 2 R 3 u ¯ L Ω u d x 1 2 p R 3 ( I α * u p ) u p d x 1 4 R 3 u 2 d x + 1 4 R 3 x 2 u 2 d x 1 2 R 3 u ¯ L Ω u d x u W 2 1 , 2 ( R 3 , C ) 2 ν r .

Note that ν = Ω 2 ( 1 Ω 2 ) 8 and μ = Ω . Then, we know

1 Ω 2 4 > 2 ν μ .

Moreover, if 0 < c < p 1 Ω 2 4 2 ν μ C α C 6 p 3 + α 2 p r 3 p α 5 2 2 3 + α p , then we know

1 2 μ r 1 Ω 2 4 1 p C α C 6 p 3 + α 2 p r 3 p α 5 2 c 3 + α p 2 > ν r .

We choose

c 0 c 0 ( r , Ω , p , α ) = min Ω + 1 6 r , Ω 2 ( 1 Ω 2 ) 24 , 1 Ω 2 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 2 3 + α p , p ( 1 Ω ) ( 1 Ω 2 ) 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 2 3 + α p .

Then, for all c < c 0 , we obtain

inf u S c B ν r J ( u ) ν r < 1 2 μ r 1 Ω 2 4 1 p C α C 6 p 3 + α 2 p r 3 p α 5 2 c 3 + α p 2 inf u S c ( B r B μ r ) J ( u ) .

The proof is completed.□

Proof of Theorem 1.3

We divide this proof into three steps.

Step 1. In this step, we show the existence of a local minimizer to equation ( C ) on S c B r . Let { u n } S c B r be a minimizing sequence for

m c , r = inf u S c B r J ( u ) .

Then, { u n } is bounded in W 2 1 , 2 ( R 3 , C ) . From Lemma 2.1, there exists u W 2 1 , 2 ( R 3 , C ) such that

u n u in W 2 1 , 2 ( R 3 , C ) , u n u in L q ( R 3 , C ) , q [ 2 , 2 * ) , u n u in R 3 .

Hence, we know u S c B r .

Applying Lemma 2.4, one has that J is weakly lower semi-continuous. Therefore, we obtain

J ( u ) lim n J ( u n ) = m c , r J ( u ) ,

which shows

J ( u ) = m c , r ,

and

u n u in W 2 1 , 2 ( R 3 , C ) .

Hence, any minimizing sequence of m c , r is precompact and c , r .

Let u c c , r be the minimizer of m c , r . To show that u c is a critical point of J S c , we need to prove that u c is not on the boundary of S c B r . With the help of Lemma 5.2, for any c ( 0 , c 0 ) , we know

inf u S c B ν r J ( u ) < inf u S c ( B r B μ r ) J ( u ) ,

which gives

u c S c B r ,

where B r { u W 2 1 , 2 ( R 3 , C ) u W 2 1 , 2 ( R 3 , C ) 2 = r } . Then, u c is a critical point of J S c . So, there exists a Lagrange multiplier ω c R such that ( u c , ω c ) is a couple of weak solution to equation ( C ).

Step 2. We estimate the bound of the Lagrange multiplier ω c in this step. Note that the first eigenvalue of equation (4.3) is λ 0 = 3 , and the first eigenfunction φ 0 S 1 W 2 1 , 2 ( R 3 , C ) . Then, φ c φ 0 B r if c r 3 . It follows from Lemma 2.4 and φ is real-valued that

(5.1) m c , r J ( φ ) = 1 4 φ W 2 1 , 2 ( R 3 , C ) 2 1 2 R 3 φ ¯ L Ω φ d x 1 2 p R 3 ( I α * φ p ) φ p d x 1 2 φ W 2 1 , 2 ( R 3 , C ) 2 1 2 p R 3 ( I α * φ p ) φ p d x < 1 2 φ W 2 1 , 2 ( R 3 , C ) 2 = 3 2 c .

Note that ( u c , ω c ) c , r × R weakly solves equation ( C ). From equation (5.1), we have

c ω c = ω c u c L 2 ( R 3 , R ) 2 = 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x R 3 ( I α * u c p ) u c p d x = 2 J ( u c ) + 1 p p R 3 ( I α * u c p ) u c p d x < 2 J ( u c ) = 2 m c , r < 3 c ,

which implies

ω c < 3 .

On the other hand, by applying Lemma 2.4 and equation (4.1), one deduces

ω c u c L 2 ( R 3 , R ) 2 = 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x R 3 ( I α * u c p ) u c p d x 1 Ω 2 4 u c W 2 1 , 2 ( R 3 , C ) 2 C α C 6 p 3 + α 2 p u c W 2 1 , 2 ( R 3 , C ) 3 p α 3 c 3 + α p 2 = u c W 2 1 , 2 ( R 3 , C ) 2 1 Ω 2 4 C α C 6 p 3 + α 2 p u c W 2 1 , 2 ( R 3 , C ) 3 p α 5 c 3 + α p 2 3 u c L 2 ( R 3 , R ) 2 1 Ω 2 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 c 3 + α p 2 ,

which gives

ω c 3 1 Ω 2 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 c 3 + α p 2 ,

where c < c 0 1 Ω 2 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 2 3 + α p , see Lemma 5.2.

Step 3. In this step, we estimate the mass collapse behavior of u c as c 0 + . For any u c , r , we write u = u 1 + i u 2 , where u 1 is the real part and u 2 is the imaginary part of u . From equation (4.7), we have

c = R 3 u 2 d x = k = 0 l k l ¯ k R 3 ψ k 2 d x = k = 0 l k 2 ,

where l k = R 3 u φ k d x and l ¯ k is the conjugate of l k .

Note that u c , r B r . It follows from Lemma 2.4 and equations (5.1) and (4.1) that

3 2 c > J ( u ) = 1 4 u W 2 1 , 2 ( R 3 , C ) 2 1 2 R 3 u ¯ L Ω u d x 1 2 p R 3 ( I α * u p ) u p d x 1 Ω 2 8 u W 2 1 , 2 ( R 3 , C ) 2 1 2 p C α C 6 p 3 + α 2 p u W 2 1 , 2 ( R 3 , C ) 3 p α 3 c 3 + α p 2 1 Ω 2 8 u W 2 1 , 2 ( R 3 , C ) 2 1 2 p C α C 6 p 3 + α 2 p r 3 p α 3 c 3 + α p 2 = 1 Ω 2 8 k = 0 λ k l k 2 1 2 p C α C 6 p 3 + α 2 p r 3 p α 3 c 3 + α p 2 ,

which gives

3 k = 0 l k 2 = λ 0 k = 0 l k 2 k = 0 λ k l k 2 12 1 Ω 2 c + 4 p ( 1 Ω 2 ) C α C 6 p 3 + α 2 p r 3 p α 3 c 3 + α p 2 .

Thus, one has

u l 0 φ 0 W 2 1 , 2 ( R 3 , C ) 2 = k = 1 l k φ k W 2 1 , 2 ( R 3 , C ) 2 = k = 1 λ k l k 2 k = 0 λ k l k 2 12 1 Ω 2 c + 4 p ( 1 Ω 2 ) C α C 6 p 3 + α 2 p r 3 p α 3 c 3 + α p 2

and

u l 0 φ 0 L 2 ( R 3 , R ) 2 = k = 1 l k φ k L 2 ( R 3 , R ) 2 = k = 1 l k 2 k = 0 l k 2 4 1 Ω 2 c + 4 3 p ( 1 Ω 2 ) C α C 6 p 3 + α 2 p r 3 p α 3 c 3 + α p 2 .

Combining the above two results, we have

sup u c , r u l 0 φ 0 Σ 2 4 4 1 Ω 2 c + 4 3 p ( 1 Ω 2 ) C α C 6 p 3 + α 2 p r 3 p α 3 c 3 + α p 2 .

Then, we obtain the desired result sup u c , r u l 0 ψ 0 Σ 2 = O ( c + c 3 + α p 2 ) .□

6 Proof of Theorems 1.4 and 1.5

In this section, we prove Theorems 1.4 and 1.5.

Proof of Theorem 1.4

We show that u c is also a normalized ground-state solution for c > 0 sufficiently small. By way of contradiction, we assume that there exists v S c such that

( J S c ) ( v ) = 0 and J ( v ) < m c , r .

Note that ( J S c ) ( v ) = 0 . Then, v satisfies

1 2 Δ v + 1 2 x 2 v L Ω v = ( I α * v p ) v p 2 v + ω v , x R 3 ,

for some ω . Then by using (4.6) and (4.8), we have ω 0 and

(6.1) v W 2 1 , 2 ( R 3 , C ) 2 < 4 ( 1 Ω 2 ) ( p 1 ) 2 p J ( v ) .

It follows from equation (6.1), J ( v ) < m c , r , equation (5.1) and p > 1 that

v W 2 1 , 2 ( R 3 , C ) 2 < 4 ( 1 Ω 2 ) ( p 1 ) 2 p J ( v ) < 4 ( 1 Ω 2 ) ( p 1 ) 2 p m c , r < 12 p ( 1 Ω 2 ) ( p 1 ) c 0 , as c 0 + .

This shows that v B r for c small enough. Then, we have J ( v ) m c , r , which contradicts J ( v ) < m c , r .

Moreover, let ( u c , ω c ) c , r × R be given in Theorem 1.3. It follows from equation (6.1), J ( u c ) = m c , r , and (5.1) that

(6.2) u c W 2 1 , 2 ( R 3 , C ) 2 < 12 p ( 1 Ω 2 ) ( p 1 ) c 0 , as c 0 + .

This shows that u c 0 in W 2 1 , 2 ( R 3 , C ) as c 0 + .

Step 3. We study the asymptotic behavior of m c , r c as c 0 + . Before that, we establish the following inequality, which plays a key role in this step:

(6.3) m c , r c < m c ¯ , r c ¯ ,

where 0 < c ¯ < c < min { 1 , c 0 } ( c 0 is given in Theorem 1.3), 5 + α 3 < p < 3 + α and 0 < Ω < 1 .

Let { u n } S c ¯ B r be a bounded minimizing sequence of m c ¯ , r . By Lemma 5.2 and c ¯ < c < min { 1 , c 0 } , we can assume that { u n } B c ¯ r for n large. Then, for any c c ¯ > 1 , we have

c c ¯ R 3 u n 2 d x = c c c ¯ u n S c ,

and

c c ¯ u n W 2 1 , 2 ( R 3 , C ) 2 = c c ¯ u n W 2 1 , 2 ( R 3 , C ) 2 c c ¯ c ¯ r = c r c c ¯ u n u n B c r ,

which gives

c c ¯ u n S c B c r S c B r .

Then, we have

J c c ¯ u n c c ¯ J ( u n ) = 1 2 p c c ¯ c c ¯ p R 3 ( I α * u n p ) u n p d x .

From Step 1 of Theorem 1.3, we know that there exists C > 0 such that

lim n R 3 ( I α * u n p ) u n p d x C > 0 .

Note that c c ¯ > 1 and p > 3 + α 3 > 1 . Then, one has

J c c ¯ u n < c c ¯ J ( u n ) .

It follows from m c , r J c c ¯ u n and m c ¯ , r = J ( u n ) that

m c , r J c c ¯ u n < c c ¯ J ( u n ) = c c ¯ m c ¯ , r .

Then, we have the inequality (6.3).

It is easy to see that

2 m c , r + 1 p p R 3 ( I α * u c p ) u c p d x 2 c = 2 J ( u c ) + 1 p p R 3 ( I α * u c p ) u c p d x 2 u c L 2 ( R 3 , R ) 2 = 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x R 3 ( I α * u c p ) u c p d x 2 u c L 2 ( R 3 , R ) 2 = ω c 2 ,

which gives

(6.4) m c , r c + 1 p p R 3 ( I α * u c p ) u c p d x 2 c = ω c 2 .

It follows from equation (6.4) and Theorem 1.3 (ii) that

3 2 > m c , r c m c , r c + 1 p p R 3 ( I α * u c p ) u c p d x 2 c = ω c 2 3 2 1 Ω 2 4 C α C 6 p 3 + α 2 p r 3 p α 5 2 c 3 + α p 2 > 3 ( 1 Ω 2 ) 16 , for c small .

This shows that m c , r c has lower and upper bounds for c small. Inequality (6.3) implies that m c c is monotone decreasing. Then, by using the monotone-bounded convergence theorem, we know that the limitation lim c 0 + m c , r c exists. Thus,

lim c 0 + m c , r c 3 ( 1 Ω 2 ) 16 , 3 2 .

Then, there exist ω ¯ ¯ 3 ( 1 Ω 2 ) 8 , 3 such that

lim c 0 + m c , r c = ω ¯ ¯ 2 .

Step 4. We study the asymptotic behavior of ω c as c 0 + . It follows from equation (6.2) and c < 1 that

u c W 2 1 , 2 ( R 3 , C ) 2 < 12 p ( 1 Ω 2 ) ( p 1 ) c .

From equation (4.1) and above result, one has

0 R 3 ( I α * u c p ) u c p d x 2 c C u c W 2 1 , 2 ( R 3 , C ) 3 p α 3 c 1 + α p 2 C c 3 p α 3 2 c 1 + α p 2 C c p 1 0 , as c 0 + .

By using the Sandwich theorem, we know that lim c 0 + R 3 ( I α * u c p ) u c p d x c = 0 . Putting this result with equation (6.4), we have

0 ω c 2 m c c 1 p p R 3 ( I α * u c p ) u c p d x 2 c 0 , as c 0 + .

This shows

lim c 0 + ω c 2 = lim c 0 + m c , r c = ω ¯ ¯ 2 .

Then,

(6.5) ω ¯ ¯ = lim c 0 + ω c = lim c 0 + 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x R 3 ( I α * u c p ) u c p d x c = lim c 0 + 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x c .

It follows from Lemma 3.1 that

0 = P ( u c ) c = 1 2 R 3 u c 2 d x R 3 x 2 u c 2 d x c + 3 + α 3 p 2 p R 3 ( I α * u c p ) u c p d x c = 1 2 R 3 u c 2 d x R 3 x 2 u c 2 d x c , as c 0 + ,

which shows

(6.6) lim c 0 + R 3 u c 2 d x c = lim c 0 + R 3 x 2 u c 2 d x c .

Combining equations (6.5) and (6.6), we have

ω ¯ ¯ = lim c 0 + ω c = lim c 0 + 1 2 u c W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ c L Ω u c d x c = lim c 0 + R 3 u c 2 d x R 3 u ¯ c L Ω u c d x c = lim c 0 + R 3 x 2 u c 2 d x R 3 u ¯ c L Ω u c d x c .

We obtain the desired result.□

Lemma 6.1

Let 5 + α 3 < p < 3 + α and 0 < Ω < 1 . For c > 0 small, then the corresponding solution to equation ( C ) exists globally in time.

Proof

Let u t ( x ) u ( t , x ) : ( T * , T * ) × R N C be the corresponding solution to equation ( C ). From c > 0 small, there exists r > 0 such that

u 0 W 2 1 , 2 ( R 3 , C ) 2 r ,

where u 0 ( x ) u ( 0 , x ) . Applying equation (4.3), we have

(6.7) R 3 u 0 2 d x 1 3 u 0 W 2 1 , 2 ( R 3 , R ) 2 1 3 u 0 W 2 1 , 2 ( R 3 , C ) 2 r 3 .

It follows from Lemma 2.4 and equations (4.1) and (6.7) that

(6.8) J ( u 0 ) 1 4 u 0 W 2 1 , 2 ( R 3 , C ) 2 + 1 2 p R 3 ( I α * u 0 p ) u 0 p d x 1 4 u 0 W 2 1 , 2 ( R 3 , C ) 2 + C u 0 W 2 1 , 2 ( R 3 , C ) 3 p α 3 u 0 L 2 ( R 3 , R ) 3 + α p C r ,

for some constant C r > 0 depending on r .

By Lemma 2.4 and equation (4.1), and the mass and energy conservations, for all t ( T * , T * ) , we obtain

1 Ω 2 4 u t W 2 1 , 2 ( R 3 , C ) 2 1 2 u t W 2 1 , 2 ( R 3 , C ) 2 R 3 u ¯ t L Ω u t d x = 2 J ( u t ) + 1 p R 3 ( I α * u t p ) u t p d x 2 J ( u t ) + C p u t W 2 1 , 2 ( R 3 , C ) 3 p α 3 u t L 2 ( R 3 , R ) 3 + α p 2 J ( u 0 ) + C p u t W 2 1 , 2 ( R 3 , C ) 3 p α 3 u 0 L 2 ( R 3 , R ) 3 + α p

where u t ( x ) u ( t , x ) . Moreover, we obtain

(6.9) u t W 2 1 , 2 ( R 3 , C ) 2 A + B u t W 2 1 , 2 ( R 3 , C ) 3 p α 3 , t ( T * , T * ) ,

where

A 4 1 Ω 2 J ( u 0 ) + u 0 W 2 1 , 2 ( R 3 , C ) 2 , B 4 C p ( 1 Ω 2 ) u 0 L 2 ( R 3 , R ) 3 + α p .

From p ( 5 + α 3 , 3 + α ) and equation (6.9), there exists C t > 1 (depending on t ) such that

u t W 2 1 , 2 ( R 3 , C ) 2 C t A .

We now show that T * = T * = . Note that u 0 W 2 1 , 2 ( R 3 , C ) 2 C t A and the continuity in time. By way of contradiction, we assume that there exists t * ( T * , T * ) such that u t * W 2 1 , 2 ( R 3 , C ) 2 = C t A . Putting this into equation (6.9), one has

C t A A + B ( C t A ) 3 p α 3 2 ,

which gives

(6.10) C t 1 C t 3 p α 5 2 A α + 5 3 p 2 B .

From equations (6.7) and (6.8), one deduces that A is bounded from above by a constant depending on r . By using this result, p > 5 + α 3 , and equation (6.10), we know that B is bounded from below by a constant B ¯ r depending on r , i.e.,

B B ¯ r > 0 .

However, B = 4 C p ( 1 Ω 2 ) u 0 L 2 ( R 3 , R ) 3 + α p = 4 C p ( 1 Ω 2 ) c 3 + α p 2 < B r for c > 0 small enough. We obtain a contradiction. Hence, T * = T * = .□

We are now in a position to prove Theorem 1.5.

Proof of Theorem 1.5

We prove that c , r is orbitally stable under the flow of ( C ). As in the proof of Theorem 1.2, we argue by contradiction. We suppose that the set c , r is unstable. Then, there exists ε 1 > 0 , w 0 c , r and a sequence of initial data { w n 0 } Σ satisfying

w n 0 w 0 Σ 0 , as n ,

and a sequence of times { t n } R such that

inf w c , r w n ( t n , ) w Σ > ε 0 ,

where w n ( 0 , ) = w n 0 ( ) . By Lemma 5.2, the corresponding solution to equation ( C ) exists globally in time for c > 0 small.

Without loss of generality, we may assume that { v n 0 } S c . For simplicity, set w ¯ n ( x ) w n ( t n , x ) . We have mass conservation, as n ,

R 3 w ¯ n 2 d x R 3 w n ( t n , x ) 2 d x = R 3 w n 0 ( x ) 2 d x R 3 w 0 ( x ) 2 d x = c

and energy conservation

J ( w ¯ n ) J ( w n ( t n , ) ) = J ( w n 0 ) J ( w 0 ) = m c , r .

We need to show that w n ( t n , ) W 2 1 , 2 ( R 3 , C ) 2 r . Suppose, on the contrary, that there exists n * such that w n ( t n , ) W 2 1 , 2 ( R 3 , C ) 2 > r for n > n * . From the continuity in time, there exists t n * such that w n ( t n * , ) W 2 1 , 2 ( R 3 , C ) 2 = r . Then, we have

R 3 w n ( t n * , ) 2 d x c , w n ( t n * , ) W 2 1 , 2 ( R 3 , C ) 2 = r , J ( w n ( t n * , ) ) m c , r , as n .

This gives that { w n ( t n * , ) } is a minimizing sequence of m c , r . By the proof of Theorem 1.3, there exists a subsequence such that

w n ( t n * , ) w ¯ Σ strongly , w ¯ W 2 1 , 2 ( R 3 , C ) 2 = r , J ( w ¯ ) = m c , r .

This is a contradiction, since the minimizer of m c , r does not belong to the set S c B r , see Lemma 5.2. Thus, there exists a subsequence { t n k } such that w n ( t n k , ) W 2 1 , 2 ( R 3 , C ) 2 r . By the proof of Theorem 1.3 again, there exists w ¯ ¯ Σ such that

inf w c , r w n ( t n , ) w Σ w n ( t n k , ) w ¯ ¯ Σ 0 , as n .

This is a contradiction.□

7 Proof of Theorem 1.6

In this section, we show Theorem 1.6, i.e., the existence of mountain pass-type solution. First, we construct a min-max class by using the result of Theorem 1.3.

Lemma 7.1

Let u c c , r be given in Theorem 1.3. Set v c l 3 2 u c ( l x ) with l > 1 . Then, there exists l > 1 large enough such that

v c S c \ B r a n d J ( v c ) < 0 .

Then, there exists a nonempty min-max class

Γ c { g C ( [ 0 , 1 ] , S c ) g ( 0 ) = u c , a n d g ( 1 ) = v c } ,

where g ( t ) = ( 1 + l t t ) 3 2 u c ( x + t ( l 1 ) x ) , and a min-max value

γ c inf g Γ c max t [ 0 , 1 ] J ( g ( t ) ) > 0 .

Proof

Let u c c , r be given in Theorem 1.3. Then, for fixed c > 0 , there exists C u c such that

(7.1) R 3 u c 2 d x = C u c R 3 x 2 u c 2 d x > 0 .

Set v c l 3 2 u c ( l x ) with l > 1 . Then, we have

v c L 2 ( R 3 , R ) 2 = u c L 2 ( R 3 , R ) 2 = c ,

and

(7.2) v c W 2 1 , 2 ( R 3 , C ) 2 = l 2 R 3 u c 2 d x + l 2 R 3 x 2 u c 2 d x l 2 u c W 2 1 , 2 ( R 3 , C ) 2 ,

and by using equation (7.1),

(7.3) v c W 2 1 , 2 ( R 3 , C ) 2 = l 2 R 3 u c 2 d x + l 2 R 3 x 2 u c 2 d x l 2 R 3 u c 2 d x = l 2 min 1 2 , C u c 2 u c W 2 1 , 2 ( R 3 , C ) 2 ,

and by applying equation (7.2),

(7.4) J ( v c ) = 1 4 v c W 2 1 , 2 ( R 3 , C ) 2 1 2 R 3 v ¯ c L Ω v c d x 1 2 p R 3 ( I α * v c p ) v c p d x l 2 4 u c W 2 1 , 2 ( R 3 , C ) 2 1 2 R 3 u ¯ c L Ω u c d x l 3 p 3 α 2 p R 3 ( I α * u c p ) u c p d x .

It follows from equations (7.3) and (7.4) and p > 3 + α 3 that there exists l > 1 large enough such that ( r is fixed, since u c in Theorem 1.3)

v c W 2 1 , 2 ( R 3 , C ) 2 > r and J ( v c ) < 0 .

These show that

v c S c \ B r and J ( v c ) < 0 .

Set g ( t ) = ( 1 + l t t ) 3 2 u c ( x + t ( l 1 ) x ) . We have g ( 0 ) = u c and g ( 1 ) = v c . Then, we can construct

Γ c { g C ( [ 0 , 1 ] , S c ) g ( 0 ) = u c , and g ( 1 ) = v c } ,

and a min-max value

γ c inf g Γ c max t [ 0 , 1 ] J ( g ( t ) ) .

It follows from J ( u c ) = m c , r > 0 , J ( v c ) < 0 and Lemma 5.2 that

γ c > max { J ( u c ) , J ( v c ) } > 0 .

The proof is completed.□

Second, we introduce an auxiliary functional J ¯ ( u , θ ) J ( κ ( u , θ ) ) : S c × R R , where κ ( u , θ ) = e 3 2 θ u ( e θ x ) and

J ¯ ( u , θ ) J ( κ ( u , θ ) ) = e 2 θ 4 R 3 u 2 d x + e 2 θ 4 R 3 x 2 u 2 d x 1 2 R 3 u ¯ L Ω u d x e ( 3 p 3 α ) θ 2 p R 3 ( I α * u p ) u p d x .

Define a set of paths

Γ ¯ c { g ¯ C ( [ 0 , 1 ] , S c × R ) g ¯ ( 0 ) = ( u c , 0 ) , and g ¯ ( 1 ) = ( v c , 0 ) } ,

and a min-max value

γ ¯ c inf g ¯ Γ ¯ c max t [ 0 , 1 ] J ¯ ( g ¯ ( t ) ) > 0 .

We need to show that γ ¯ c = γ c . It follows from the definition of γ c and γ ¯ c , and the following maps

ϕ : Γ c Γ ¯ c , ϕ ( g ) = ( g , 0 ) and Φ : Γ ¯ c Γ c , Φ ( g ¯ ) = κ g ,

satisfy

J ¯ ( ϕ ( g ) ) = J ( g ) and J ( Φ ( g ¯ ) ) = J ¯ ( g ¯ ) .

Denote r R for r R , E Σ × R with the norm E 2 = W 2 1 , 2 ( R 3 , C ) 2 + R 2 and E 1 is the dual space of E .

Lemma 7.2

Let ε > 0 . Let g ¯ Γ ¯ c satisfies

max t [ 0 , 1 ] J ¯ ( g ¯ ( t ) ) γ ¯ c + ε .

Then, there exists a pair of ( u 0 , θ 0 ) S c × R such that

  1. J ¯ ( u 0 , θ 0 ) [ γ ¯ c ε , γ ¯ c + ε ] ;

  2. min t [ 0 , 1 ] ( u 0 , θ 0 ) g ¯ ( t ) E ε ;

  3. for all z T ¯ ( u 0 , θ 0 ) { ( z 1 , z 2 ) E u 0 , z 1 L 2 ( R 3 ) = 0 } , one has

    ( J ¯ S c × R ) ( u 0 , θ 0 ) E 1 2 ε , i.e. , J ¯ ( u 0 , θ 0 ) , z 2 ε z E .

Proof

The proof is similar to that of [23, Lemma 2.3].□

Definition 7.1

We say the { v n } S c is a Pohozaev-type Palais-Smale sequence for γ c (short for ( PPS ) γ c ) if it satisfies

J ( v n ) γ c , ( J S c ) ( v n ) 0 , P ( v n ) 0 .

We construct a ( PPS ) γ c sequence in next lemma.

Lemma 7.3

Let α ( 0 , 3 ) , 5 + α 3 < p < 3 + α , 0 < Ω < 1 , and c < c 0 ( c 0 is given in Theorem 1.3). Then, there exists a ( PPS ) γ c sequence { v n } S c .

Proof

For each n N + , there exists g n Γ c such that

max t [ 0 , 1 ] J ( g n ( t ) ) γ c + 1 n .

Note that γ ¯ c = γ c and g ¯ n = ( g n , 0 ) Γ ¯ c . Then, we have

max t [ 0 , 1 ] J ¯ ( g ¯ n ( t ) ) γ ¯ c + 1 n .

It follows from Lemma 7.2 that there exists a sequence { ( u n , θ n ) } S c × R such that

  1. J ¯ ( u n , θ n ) [ γ ¯ c 1 n , γ ¯ c + 1 n ] ;

  2. min t [ 0 , 1 ] ( u n , θ n ) g ¯ n ( t ) E 1 n ;

  3. for all z T ¯ ( u n , θ n ) { ( z 1 , z 2 ) E u n , z 1 L 2 ( R 3 ) = 0 } , one has

    ( J ¯ S c × R ) ( u n , θ n ) E 1 2 1 n , i.e. , J ¯ ( u n , θ n ) , z 2 1 n z E .

Set v n = κ ( u n , θ n ) . We now show that { v n } S c is a ( PPS ) γ c sequence. First, from equation ( 1 ) , we have

J ( v n ) = J ( κ ( u n , θ n ) ) = J ¯ ( u n , θ n ) γ ¯ c = γ c , as n .

It is easy to see that

2 P ( v n ) = R 3 v n 2 d x R 3 x 2 v n 2 d x + 3 + α 3 p p R 3 ( I α * v n p ) v n p d x = e 2 θ n R 3 u n 2 d x e 2 θ n R 3 x 2 u n 2 d x 3 p 3 α p e ( 3 p 3 α ) θ n R 3 ( I α * v n p ) v n p d x = J ¯ ( u n , θ n ) , ( 0 , 1 ) .

From ( 3 ) , we obtain P ( v n ) 0 as n , for ( 0 , 1 ) T ¯ ( u n , θ n ) .

We now show that, for n large enough,

J ( v n ) , η 4 1 n η Σ , η T v n { η Σ v n , η L 2 ( R 3 ) = 0 } .

For any η T v n , let η ¯ κ ( η , θ n ) , we have

J ( v n ) , η = J ¯ ( u n , θ n ) , ( η ¯ , 0 ) .

Note that R 3 u n η ¯ d x = R 3 v n η d x . Then, we obtain ( η ¯ , 0 ) T ¯ u n , θ n η T v n . By using equation (2), one has

θ n = θ n 0 min t [ 0 , 1 ] ( u n , θ n ) g ¯ n ( t ) 1 n .

For n large enough, we obtain

( η ¯ , 0 ) E 2 = η ¯ Σ 2 + 0 R 2 = e 2 θ n η D 1 , 2 ( R 3 , C ) 2 + e 2 θ n x η L 2 ( R 3 , R ) 2 + η L 2 ( R 3 , R ) 2 2 η Σ 2 .

From equation (3), we obtain

J ( v n ) , η = J ¯ ( u n , θ n ) , ( η ¯ , 0 ) 2 1 n ( η ¯ , 0 ) E 4 1 n η Σ .

Then,

( J S c ) ( v n ) Σ 1 = sup η T v n , η Σ 1 J ( v n ) , η 4 1 n 0 , as n .

The proof is completed.□

Lemma 7.4

Let { v n } S c be a bounded ( PPS ) γ c sequence. Then, there exists λ ¯ n = J ( v n ) , v n c such that

( J S c ) ( v n ) 0 , i n Σ 1 J ( v n ) λ ¯ n v n 0 , in Σ 1 , a s n .

Proof

The proof is similar to [6, Lemma 3], we omit it.□

Lemma 7.5

Let α ( 0 , 3 ) , 5 + α 3 < p < 3 + α , 0 < Ω < 1 , and c < c 0 ( c 0 is given in Theorem 1.3). Let { v n } S c be a ( PPS ) γ c sequence. Then, there exist a v W 2 1 , 2 ( R 3 , C ) , a sequence { ω n } R , and a ω ˜ R such that

  1. up to a subsequence, v n v in W 2 1 , 2 ( R 3 , C ) as n ;

  2. up to a subsequence, ω n ω ˜ in R as n ;

  3. up to a subsequence, Re [ J ( v n ) ω n v n ] = 0 in Σ 1 as n ;

  4. Re [ J ( v ) ω ˜ v ] = 0 in Σ 1 .

Proof

It follows from Lemma 7.4 that

( J S c ) ( v n ) 0 , in Σ 1 J ( v n ) 1 c J ( v n ) , v n v n 0 , in Σ 1 , as n .

Then, we have

Re J ( v n ) 1 c J ( v n ) , v n v n , φ 0 , as n ,

for any φ Σ , i.e.,

Re 1 2 R 3 v n φ d x + 1 2 R 3 x 2 v n φ d x R 3 φ ¯ L Ω v n d x R 3 ( I α * v n p ) v n p 2 v n φ d x ω n R 3 v n φ d x 0 , as n ,

where

(7.5) ω n = 1 c J ( v n ) , v n = 1 c 1 2 v n W 2 1 , 2 ( R 3 , C ) 2 R 3 v ¯ n L Ω v n d x R 3 ( I α * v n p ) v n p d x .

Thus, (3) is proved.

In order to prove equation (2), we need to show that { ω n } is bounded in R . It follows from equation (7.5) and Lemma J ( v n ) < γ c + 1 that

ω n = 1 c 1 2 v n W 2 1 , 2 ( R 3 , C ) 2 R 3 v ¯ n L Ω v n d x R 3 ( I α * v n p ) v n p d x = 1 c 2 J ( v n ) + 1 p p R 3 ( I α * v n p ) v n p d x 2 c J ( v n ) < 2 ( γ c + 1 ) c .

This shows that { ω n } is bounded in R . Up to a subsequence, there exists ω ˜ R such that

ω n ω ˜ in R as n .

Thus, equation (2) is proved.

We show that { v n } is bounded in W 2 1 , 2 ( R 3 , C ) . It follows from the definition of ( PPS ) γ c sequence that

1 + γ c + o n ( 1 ) J ( v n ) + o n ( 1 ) = J ( v n ) 1 2 p J ( v n ) , v n = 1 2 1 2 1 2 p v n W 2 1 , 2 ( R 3 , C ) 2 2 R 3 v ¯ n L Ω v n d x ω n 2 p R 3 v n 2 d x 1 2 p 1 p 1 Ω 2 4 v n W 2 1 , 2 ( R 3 , C ) 2 ω n 2 p R 3 v n 2 d x ,

which shows

1 2 p 1 p 1 Ω 2 4 v n W 2 1 , 2 ( R 3 , C ) 2 1 + γ c + ω n 2 p R 3 v n 2 d x + o n ( 1 ) 1 + γ c + c 2 p ω n + o n ( 1 ) .

Note that { ω n } is bounded in R . This shows that { v n } is bounded in W 2 1 , 2 ( R 3 , C ) . Up to a subsequence, there exists v W 2 1 , 2 ( R 3 , C ) such that

v n v in W 2 1 , 2 ( R 3 , C ) , v n v in R 3 .

Moreover, combining equations (2) and (3), we have equation (4).

Finally, by using (2)–(4), we have

(7.6) Re J ( v n ) ω n v n , v n v = o n ( 1 ) , and Re J ( v ) ω ˜ v , v n v = 0 .

According to equation (7.6) and Lemma 2.4, one has

o n ( 1 ) = 1 2 v n v W 2 1 , 2 ( R 3 , C ) 2 R 3 ( v n v ) ¯ L Ω ( v n v ) d x 1 Ω 2 4 v n v W 2 1 , 2 ( R 3 , C ) 2 .

Then, equation (1) is proved.□

We are now in a position to show Theorem 1.6.

Proof of Theorem 1.6

From Lemma 7.3, we know that there exists a ( PPS ) γ c sequence { v n } S c . Furthermore, Lemma 7.5 guarantees the existence of weak solution ( u ¯ c , ω ¯ c ) Σ × R to equation ( C ), where

R 3 u ¯ c 2 d x = c

and

J ( u ¯ c ) = γ c > m c , r = J ( u c ) .

The proof is completed.□

  1. Funding information: This paper is supported by the National Natural Science Foundation of China (Grant No. 12101006).

  2. Author contributions: All authors have accepted responsibility for the entire content of this manuscript and consented to its submission to the journal, reviewed all the results and approved the final version of the manuscript. The authors contributed equally to this study.

  3. Conflict of interest: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.

  4. Data availability statement: No data were used for the research described in the article.

References

[1] P. Antonelli, R. Carles, and J. Silva, Scattering for nonlinear Schrödinger equation under partial harmonic confinement, Comm. Math. Phys. 334 (2015), 367–396, DOI: https://doi.org/10.1007/s00220-014-2166-y. Search in Google Scholar

[2] J. Arbunich, I. Nenciu, and C. Sparber, Stability and instability properties of rotating Bose-Einstein condensates, Lett. Math. Phys. 109 (2019), no. 6, 1415–1432, DOI: https://doi.org/10.1007/s11005-018-01149-5. Search in Google Scholar

[3] A. Ardila and H. Hajaiej, Global well-posedness, blow-up and stability of standing waves for supercritical NLS with rotation, J. Dynam. Differential Equations 35 (2021), 1643–1665, DOI: https://doi.org/10.1007/s10884-021-09976-2. Search in Google Scholar

[4] J. Bellazzini, N. Boussaid, L. Jeanjean, and N. Visciglia, Existence and stability of standing waves for supercritical NLS with a partial confinement, Commun. Math. Phys. 353 (2017), 229–251, DOI: https://doi.org/10.1007/s00220-017-2866-1. Search in Google Scholar

[5] J. Bellazzini and L. Jeanjean, On dipolar quantum gases in the unstable regime, SIAM J. Math. Anal. 48 (2016), 2028–2058, DOI: https://doi.org/10.1137/15M1015959. Search in Google Scholar

[6] H. Berestycki and P. Lions, Nonlinear scalar field equations, II existence of infinitely many solutions, Arch. Ration. Mech. Anal. 82 (1983), 347–375, DOI: https://doi.org/10.1007/BF00250556. Search in Google Scholar

[7] D. Cao, B. Feng, and T. Luo, On the standing waves for the X-ray free electron laser Schrödinger equation, Discrete Contin. Dyn. Syst. 42 (2022), no. 12, 6097–6137, DOI: https://doi.org/10.3934/dcds.2022139. Search in Google Scholar

[8] R. Carles and H. Hajaiej, Complementary study of the standing wave solutions of the Gross-Pitaevskii equation in dipolar quantum gases, Bull. Lond. Math. Soc. 47 (2015), no. 3, 509–518, DOI: https://doi.org/10.1112/blms/bdv024. Search in Google Scholar

[9] S. Cingolani and K. Tanaka, Semi-classical states for the nonlinear Choquard equations: existence, multiplicity and concentration at a potential well, Rev. Mat. Iberoam. 35 (2019), no. 6, 1885–1924. DOI: https://doi.org/10.4171/rmi/1105. Search in Google Scholar

[10] S. Cingolani, M. Gallo, and K. Tanaka, Multiple solutions for the nonlinear Choquard equation with even or odd nonlinearities, Calc. Var. Partial Differential Equations 61 (2022), 68. Search in Google Scholar

[11] S. Cingolani and K. Tanaka, Ground state solutions for the nonlinear Choquard equation with prescribed mass, in: V.Ferone, T.Kawakami, P.Salani, F. Takahashi, (Eds.), Geometric Properties for Parabolic and Elliptic PDE’s, Springer, Cham, 2021, pp. 23–41. Search in Google Scholar

[12] Y. Ding and X. Zhong, Normalized solution to the Schrödinger equation with potential and general nonlinear term: Mass super-critical case, J. Differential Equations 334 (2022), 194–215, DOI: https://doi.org/10.1016/j.jde.2022.06.013. Search in Google Scholar

[13] V. Dinh, Existence and stability of standing waves for nonlinear Schrödinger equation with a critical rotation speed, Lett. Math. Phys. 112 (2022), 53, 36pp. Search in Google Scholar

[14] V. Dinh, Remarks on nonlinear Schrödinger equations arising on rotational Bose-Einstein condensates, Nonlinear Anal. 214 (2022), 112587, 37pp. DOI: https://doi.org/10.1016/j.na.2021.112587. Search in Google Scholar

[15] V. Dinh, L. Forcella, and H. Hajaiej, Mass-energy threshold dynamics for dipolar quantum gases, Commun. Math. Sci. 20 (2022), no. 1, 165–200, DOI: https://doi.org/10.4310/CMS.2022.v20.n1.a5. Search in Google Scholar

[16] L. Diósi, Gravitation and quantum-mechanical localization of macro-objects, Phys. Lett. A 105 (1984), 199–202, DOI: https://doi.org/10.1016/0375-9601(84)90397-9. Search in Google Scholar

[17] L. Du, F. Gao, and M. Yang, On elliptic equations with Stein-Weiss type convolution parts, Math. Z. 301 (2022), 1–41, DOI: https://doi.org/10.1007/s00209-022-02973-1. Search in Google Scholar

[18] L. Erdos, B. Schlein, and H. Yau, Derivation of the Gross-Pitaevskii hierarchy for the dynamics of Bose-Einstein condensate, Comm. Pure Appl. Math. 59 (2006), no. 12, 1659–1741, DOI: https://doi.org/10.1002/cpa.20123. Search in Google Scholar

[19] F. Gao and M. Yang, The Brezis-Nirenberg type critical problem for the nonlinear Choquard equation, Sci. China Math. 61 (2018), no. 7, 1219–1242, DOI: https://doi.org/10.1007/s11425-016-9067-5. Search in Google Scholar

[20] J. Giacomoni, D. Goel, and K. Sreenadh, Regularity results on a class of doubly nonlocal problems, J. Differential Equations 268 (2020), 5301–5328, DOI: https://doi.org/10.1016/j.jde.2019.11.009. Search in Google Scholar

[21] Y. Guo, Y. Luo, and S. Peng, Local uniqueness of ground-states for rotating Bose-Einstein condensates with attractive interactions, Calc. Var. Partial Differential Equations 60 (2021), no. 6, Paper No. 237, 27pp, DOI: https://doi.org/10.1007/s00526-021-02055-w. Search in Google Scholar

[22] Y. Guo, Y. Luo, and W. Yang, The nonexistence of vortices for rotating Bose-Einstein condensates with attractive interactions, Arch. Ration. Mech. Anal. 238 (2020)1231–1281, DOI: https://doi.org/10.1007/s00205-020-01564-w. Search in Google Scholar

[23] L. Jeanjean, Existence of solutions with prescribed norm for semilinear elliptic equations, Nonlinear Anal. 28 (1997), 1633–1659, DOI: https://doi.org/10.1016/S0362-546X(96)00021-1. Search in Google Scholar

[24] L. Jeanjean, J. Jendrej, T. Le, and N. Visciglia, Orbital stability of ground-states for a Sobolev critical Schrödinger equation, J. Math. Pures Appl. 164 (2022), 158–179, DOI: https://doi.org/10.1016/j.matpur.2022.06.005. Search in Google Scholar

[25] L. Jeanjean and T. Le, Multiple normalized solutions for a Sobolev critical Schrödinger-Poisson-Slater equation, J. Differential Equations, 303 (2021), 277–325, DOI: https://doi.org/10.1016/j.jde.2021.09.022. Search in Google Scholar

[26] M. Lewin, P. Nam, and N. Rougerie, The mean-field approximation and the non-linear Schrödinger functional for trapped Bose gases, Trans. Amer. Math. Soc. 368 (2016), no. 9, 6131–6157, DOI: https://doi.org/10.1090/tran/6537. Search in Google Scholar

[27] M. Lewin, P. Nam, and N. Rougerie, Derivation of Hartree’s theory for generic mean-field Bose systems, Adv. Math. 254 (2014), 570–621, DOI: https://doi.org/10.1016/j.aim.2013.12.010. Search in Google Scholar

[28] G. Li and H. Ye, The existence of positive solutions with L2-norm for nonlinear Choquard equation, J. Math. Phys. 55 (2014), 121501, DOI: https://doi.org/10.1063/1.4902386. Search in Google Scholar

[29] X. Li and S. Ma, Ground states for Choquard equations with doubly critical exponents, Rocky Mountain J. Math. 49 (2019), no. 1, 153–170, DOI: https://doi.org/10.1216/RMJ-2019-49-1-153. Search in Google Scholar

[30] Y. Li, G. Li, and C. Tang, Existence and concentration of ground-state solutions for Choquard equations involving critical growth and steep potential well, Nonlinear Anal. 200 (2020), no. 1, 111997, DOI: https://doi.org/10.1016/j.na.2020.111997. Search in Google Scholar

[31] E. Lieb, R. Seiringer, and J. Yngvason, Bosons in a trap: a rigorous derivation of the Gross-Pitaevskii energy functional, Phys. Rev. A 61 (2000), 043602, DOI: https://doi.org/10.1103/PhysRevA.61.043602. Search in Google Scholar

[32] E. Lieb, R. Seiringer, and J. Yngvason, A rigorous derivation of the Gross-Pitaevskii energy functional for a two-dimensional Bose gas, Comm. Math. Phys. 224 (2001), 17–31, DOI: https://doi.org/10.1007/s002200100533. Search in Google Scholar

[33] E. Lieb, Existence and uniqueness of the minimizing solution of Choquardas nonlinear equation, Studies Appl. Math. 57 (1976/77), no. 2, 93–105, DOI: https://doi.org/10.1002/sapm197757293. Search in Google Scholar

[34] E. Lieb and M. Loss, Analysis, Graduate Studies in Mathematics, vol. 14, American Mathematical Society, Providence, RI, 1997. Search in Google Scholar

[35] P. Lions, The Choquard equation and related questions, Nonlinear Anal. 4 (1980), no. 6, 1063–1072, DOI: https://doi.org/10.1016/0362-546X(80)90016-4. Search in Google Scholar

[36] P. Lions, The concentration-compactness principle in the calculus of variations, the locally compact case, part 1, Ann. Inst. H. Poincaré Non Lineaire 1 (1984), 109–145, DOI: https://doi.org/10.1016/S0294-1449(16)30428-0. Search in Google Scholar

[37] X. Luo and T. Yang, Multiplicity, asymptotics and stability of standing waves for nonlinear Schrödinger equation with rotation, J. Differential Equations 304 (2021), 326–347, DOI: https://doi.org/10.1016/j.jde.2021.10.003. Search in Google Scholar

[38] V. Moroz and J. Van Schaftingen, Groundstates of nonlinear Choquard equations: existence, qualitative properties and decay asymptotics, J. Funct. Anal. 265 (2013), no. 2, 153–184, DOI: https://doi.org/10.1016/j.jfa.2013.04.007. Search in Google Scholar

[39] V. Moroz and J. Van Schaftingen, Existence of groundstates for a class of nonlinear Choquard equations, Trans. Amer. Math. Soc. 367 (2015), no. 9, 6557–6579, DOI: https://doi.org/10.1090/S0002-9947-2014-06289-2. Search in Google Scholar

[40] S. Pekar, Untersuchung ber die elektronentheorie der kristalle, Akademie Verlag, Berlin, 1954, DOI: https://doi.org/10.1515/9783112649305. Search in Google Scholar

[41] R. Penrose, On gravityas role in quantum state reduction, Gen. Relativity Gravitation 28 (1996), no. 5, 581–600, DOI: https://doi.org/10.1007/BF02105068. Search in Google Scholar

[42] L. Pitaevskii and S. Stringari, Bose-Einstein Condensation, International Series of Monographs on Physics, vol. 116, The Clarendon Press, Oxford University Press, Oxford, 2003, https://global.oup.com/academic/product/bose-einstein-condensation-9780198507192?cc=cnlang=en. Search in Google Scholar

[43] L. Santos, G. Shlyapnikov, P. Zoller, and M. Lewenstein, Bose-Einstein condensation in trapped dipolar gases, Phys. Rev. Lett. 85 (2000), 1791–1797, DOI: https://doi.org/10.1103/PhysRevLett.85.1791. Search in Google Scholar

[44] R. Seiringer, Gross-Pitaevskii theory of the rotating Bose gas, Comm. Math. Phys. 229 (2002), no. 3, 491–509, DOI: https://doi.org/10.1007/s00220-002-0695-2. Search in Google Scholar

[45] F. Selem, H. Hajaiej, P. Markowich, and S. Trabelsi, Variational approach to the orbital stability of standing waves of the Gross-Pitaevskii equation, Milan J. Math. 84 (2014), 273–295, DOI: https://doi.org/10.1007/s00032-014-0227-5. Search in Google Scholar

[46] N. Soave, Normalized ground-states for the NLS equation with combined nonlinearities: The Sobolev critical case, J. Funct. Anal. 279 (2020), 108610, DOI: https://doi.org/10.1016/j.jfa.2020.108610. Search in Google Scholar

[47] Y. Su and Z. Liu, Semi-classical states to nonlinear Choquard equation with critical growth, Israel J Math. 255 (2023), 729–762. Search in Google Scholar

[48] A. Triay, Derivation of the dipolar Gross-Pitaevskii energy, SIAM J. Math. Anal. 50 (2018), 33–63, DOI: https://doi.org/10.1137/17M112378X. Search in Google Scholar

[49] M. Weinstein, Nonlinear Schrödinger equations and sharp interpolation estimates, Comm. Math. Phys. 87 (1983), 567–576, DOI: https://doi.org/10.1007/BF01208265. Search in Google Scholar

[50] M. Yang, V. Radulescu, and X. Zhou, Critical Stein-Weiss elliptic systems: symmetry, regularity and asymptotic properties of solutions, Calc. Var. Partial Differential Equations 61 (2022), 109, DOI: https://doi.org/10.1007/s00526-022-02221-8. Search in Google Scholar

[51] S. Yao, H. Chen, V. Radulescu, and J. Sun, Normalized solutions for lower critical Choquard equations with critical Sobolev perturbations, SIAM J. Math. Anal. 54 (2022), 3696–3723, DOI: https://doi.org/10.1137/21M1463136. Search in Google Scholar

Received: 2023-04-13
Revised: 2023-10-07
Accepted: 2024-01-26
Published Online: 2024-06-05

© 2024 the author(s), published by De Gruyter

This work is licensed under the Creative Commons Attribution 4.0 International License.

Articles in the same Issue

  1. Regular Articles
  2. Existence of normalized peak solutions for a coupled nonlinear Schrödinger system
  3. Orbital stability of the trains of peaked solitary waves for the modified Camassa-Holm-Novikov equation
  4. Global boundedness in a two-dimensional chemotaxis system with nonlinear diffusion and singular sensitivity
  5. Existence and uniqueness of solution for a singular elliptic differential equation
  6. The bounded variation capacity and Sobolev-type inequalities on Dirichlet spaces
  7. Vanishing and blow-up solutions to a class of nonlinear complex differential equations near the singular point
  8. Existence, uniqueness, localization and minimization property of positive solutions for non-local problems involving discontinuous Kirchhoff functions
  9. Concentration phenomena for a fractional relativistic Schrödinger equation with critical growth
  10. Beyond the classical strong maximum principle: Sign-changing forcing term and flat solutions
  11. Monotonicity of solutions for parabolic equations involving nonlocal Monge-Ampère operator
  12. On a nonlinear Robin problem with an absorption term on the boundary and L1 data
  13. Multiple solutions for the quasilinear Choquard equation with Berestycki-Lions-type nonlinearities
  14. Gradient estimates for a class of higher-order elliptic equations of p-growth over a nonsmooth domain
  15. Global existence and decay estimates of the classical solution to the compressible Navier-Stokes-Smoluchowski equations in ℝ3
  16. Global existence and finite-time blowup for a mixed pseudo-parabolic r(x)-Laplacian equation
  17. Multiple positive solutions for a class of concave-convex Schrödinger-Poisson-Slater equations with critical exponent
  18. A minimization problem with free boundary for p-Laplacian weakly coupled system
  19. Multiplicity of semiclassical solutions for a class of nonlinear Hamiltonian elliptic system
  20. k-convex solutions for multiparameter Dirichlet systems with k-Hessian operator and Lane-Emden type nonlinearities
  21. Infinitely many solutions for Hamiltonian system with critical growth
  22. On optimal control in a nonlinear interface problem described by hemivariational inequalities
  23. Variational–hemivariational system for contaminant convection–reaction–diffusion model of recovered fracturing fluid
  24. Potential and monotone homeomorphisms in Banach spaces
  25. Critical fractional Schrödinger-Poisson systems with lower perturbations: the existence and concentration behavior of ground state solutions
  26. Supercritical Hénon-type equation with a forcing term
  27. Concentration of blow-up solutions for the Gross-Pitaveskii equation
  28. Mountain-pass-type solutions for Schrödinger equations in R2 with unbounded or vanishing potentials and critical exponential growth nonlinearities
  29. Regularity for critical fractional Choquard equation with singular potential and its applications
  30. Double phase anisotropic variational problems involving critical growth
  31. Normalized solutions of NLS equations with mixed nonlocal nonlinearities
  32. Nontrivial solutions for resonance quasilinear elliptic systems
  33. Standing waves for Choquard equation with noncritical rotation
  34. Low regularity conservation laws for Fokas-Lenells equation and Camassa-Holm equation
  35. Modified quasilinear equations with strongly singular and critical exponential nonlinearity
  36. Limit profiles and the existence of bound-states in exterior domains for fractional Choquard equations with critical exponent
  37. Nests of limit cycles in quadratic systems
  38. Regularity of minimizers for double phase functionals of borderline case with variable exponents
  39. Boundedness, existence and uniqueness results for coupled gradient dependent elliptic systems with nonlinear boundary condition
  40. Ground state solutions for magnetic Schrödinger equations with polynomial growth
  41. Nonoccurrence of Lavrentiev gap for a class of functionals with nonstandard growth
  42. Dynamics for wave equations connected in parallel with nonlinear localized damping
  43. Boussinesq's equation for water waves: Asymptotics in Sector I
  44. Optimal global second-order regularity and improved integrability for parabolic equations with variable growth
  45. Normalized solutions of Schrödinger equations involving Moser-Trudinger critical growth
  46. Normalized solutions for the double-phase problem with nonlocal reaction
  47. Normalized solutions for Sobolev critical fractional Schrödinger equation
  48. Well-posedness of Cauchy problem of fractional drift diffusion system in non-critical spaces with power-law nonlinearity
  49. Improved results on planar Klein-Gordon-Maxwell system with critical exponential growth
  50. Existence and multiplicity of solutions for a new p(x)-Kirchhoff equation
  51. Quasiconvex bulk and surface energies: C1,α regularity
  52. Time decay estimates of solutions to a two-phase flow model in the whole space
  53. Bounds for the sum of the first k-eigenvalues of Dirichlet problem with logarithmic order of Klein-Gordon operators
  54. The properties of a new fractional g-Laplacian Monge-Ampère operator and its applications
  55. A fractional profile decomposition and its application to Kirchhoff-type fractional problems with prescribed mass
  56. Cauchy problem for a non-Newtonian filtration equation with slowly decaying volumetric moisture content
  57. Multiplicity of normalized semi-classical states for a class of nonlinear Choquard equations
  58. The second Yamabe invariant with boundary
  59. Peaked solitary waves and shock waves of the Degasperis-Procesi-Kadomtsev-Petviashvili equation
  60. Normalized solutions for the Choquard equations with critical nonlinearities
  61. Boundedness and long-time behavior in a parabolic-elliptic system arising from biological transport networks
  62. Initial boundary value problem and exponential stability for the planar magnetohydrodynamics equations with temperature-dependent viscosity
  63. Nonexistence of mean curvature flow solitons with polynomial volume growth immersed in certain semi-Riemannian warped products
  64. Analysis of a vector-borne disease model with vector-bias mechanism in advective heterogeneous environment
  65. Normalized solutions for the Kirchhoff equation with combined nonlinearities in ℝ4
  66. Nonexistence of the compressible Euler equations with space-dependent damping in high dimensions
  67. Multiplicity of solutions for a nonhomogeneous quasilinear elliptic equation with concave-convex nonlinearities
  68. On semilinear inequalities involving the Dunkl Laplacian and an inverse-square potential outside a ball
  69. Besov regularity for the elliptic p-harmonic equations in the non-quadratic case
  70. Uniqueness and nondegeneracy of ground states for Δ u + u = ( I α u 2 ) u in R 3 when α is close to 2
  71. Existence of solutions for a class of quasilinear Schrödinger equations with Choquard-type nonlinearity
  72. Weighted Hardy-Adams inequality on unit ball of any even dimension
  73. Existence and regularity for a p-Laplacian problem in ℝN with singular, convective, and critical reaction
  74. Temporal periodic solutions of non-isentropic compressible Euler equations with geometric effects
  75. Nodal solutions for a zero-mass Chern-Simons-Schrödinger equation
  76. The modified quasi-boundary-value method for an ill-posed generalized elliptic problem
  77. Nonlocal heat equations with generalized fractional Laplacian
  78. Choquard equations with recurrent potentials
  79. Local and global well-posedness of the Maxwell-Bloch system of equations with inhomogeneous broadening
  80. The Riemann problem for two-layer shallow water equations with bottom topography
  81. Global stability and asymptotic profiles of a partially degenerate reaction diffusion Cholera model with asymptomatic individuals
  82. On Kirchhoff-Schrödinger-Poisson-type systems with singular and critical nonlinearity
  83. Energy-variational solutions for viscoelastic fluid models
  84. Stability on 3D Boussinesq system with mixed partial dissipation
  85. Existence and multiplicity results for non-autonomous second-order Hamiltonian systems
  86. Erratum
  87. Erratum to “Normalized solutions for the Kirchhoff equation with combined nonlinearities in ℝ4
Downloaded on 10.9.2025 from https://www.degruyterbrill.com/document/doi/10.1515/anona-2023-0140/html
Scroll to top button