betaFeedback

WorksMathematics

Finite-Time Blow-up for the Radial Mass-Critical Fractional Nonlinear Schrödinger Equation

Abstract

We prove that radial negative-energy strong solutions to the focusing mass-critical fractional nonlinear Schrödinger equation i∂tu=(−Δ)su−∣u∣4s/Nu,N≥2,12<s<1, i\partial_tu=(-Δ)^su-|u|^{4s/N}u, \quad N\ge2,\quad \frac12<s<1, blow up in finite time. This resolves the finite-time blow-up versus infinite-time grow-up alternative left open by Boulenger, Himmelsbach and Lenzmann in the mass-critical regime. The argument uses an exponentially localized virial field together with a dyadic resolvent estimate that transfers the resulting weighted Balakrishnan energy to localized fractional Sobolev norms. When s≥N/(2N−1)s\ge N/(2N-1), we extend the conclusion to radial HsH^s solutions by establishing the integrated virial identity at energy regularity. As a consequence, mass-critical ground-state standing waves are strongly unstable by finite-time blow-up.

Introduction

We consider the focusing mass-critical fractional nonlinear Schrödinger equation

{i∂tu=(−Δ)su−∣u∣4sNu,u(0)=u0∈Hs(RN),(1)\begin{cases} i\partial_t u = (-\Delta)^s u - |u|^{\frac{4s}{N}}u,\\ u(0) = u_0 \in H^s(\mathbb{R}^N), \tag*{(1)} \end{cases}

where

N≥2,12<s<1.(2)N \ge2,\qquad\frac{1}{2} < s < 1. \tag*{(2)}

Throughout the paper, we write

p:=4sN,q:=p+2=2+4sN.(3)p := \frac{4s}{N},\qquad q := p+2 = 2+\frac{4s}{N}. \tag*{(3)}

Equation (1) is mass-critical since it is invariant under the mass-preserving scaling

uλ(t,x)=λN/2u(λ2st,λx).u_\lambda(t,x)=\lambda^{N/2}u(\lambda^{2s}t,\lambda x).

Along the flow, the mass

M(u):=∥u∥L22M(u):=\|u\|_{L^2}^2

and the energy

E(u):=12∥(−Δ)s/2u∥L22−1q∥u∥LqqE(u):=\frac{1}{2}\|(-\Delta)^{s/2}u\|_{L^2}^2-\frac{1}{q}\|u\|_{L^q}^q

are conserved.

The fractional Schrödinger equation was introduced by Laskin [23] in the framework of fractional quantum mechanics. Its nonlinear dispersive theory has since developed in several directions. Local well-posedness in Sobolev spaces for power-type nonlinearities was studied by Hong and Sire [22] and the second author [7], among others. In the radial setting, the improved Strichartz estimates of Guo and Wang [20] yield lossless estimates in a substantial range of fractional orders. Global well-posedness and scattering have also been investigated in critical and intercritical regimes; see, for example, Guo–Sire–Wang–Zhao [19] for the radial energy-critical equation and Sun–Wang–Yao–Zheng [28] for scattering below the ground-state threshold in the focusing case. On the elliptic side, the uniqueness and nondegeneracy of radial ground states for fractional equations were established by Frank, Lenzmann and Silvestre [15] and related works.

Our concern here is finite-time blow-up. For the classical nonlinear Schrödinger equation s=1s=1, virial and localized virial identities are among the fundamental tools for proving blow-up; see, for example, Glassey [17] and Ogawa–Tsutsumi [29]. The fractional problem is substantially more delicate because (−Δ)s(-\Delta)^s is nonlocal. In particular, the usual local virial calculus does not apply directly, and spatial localization creates interactions between different regions of space.

A fundamental advance was made by Boulenger, Himmelsbach and Lenzmann [2], who developed a localized virial method for fractional NLS based on the Balakrishnan representation of the fractional Laplacian. In the radial mass-critical case, they showed that negative energy implies the alternative

T+<∞T_{+}<\infty

or

T+=∞and∥(−Δ)s/2u(t)∥L2≳tsT_{+}=\infty\quad\text{and}\quad\lVert(-\Delta)^{s/2}u(t)\rVert_{L^2}\gtrsim t^s

for all sufficiently large tt. Thus their argument does not exclude a global solution whose fractional kinetic norm grows without bound.

The blow-up theory for fractional Schrödinger equations has since attracted a considerable interest in the mathematical community. For power-type nonlinearities, Feng [13] studied equations with combined powers, obtaining blow-up criteria, sharp global-existence/blow-up thresholds, and concentration and profile properties of blow-up solutions. In [10, 8] the second author investigated blow-up dynamics in the mass-supercritical and mass-critical regimes. He also established localized-virial blow-up criteria beyond radial data in [9]. The instability of standing waves was studied by Peng and Shi [27] in the mass-critical setting and by the second author [11] in the mass-supercritical case.

Related blow-up phenomena have also been studied for fractional Schrödinger equations with nonlocal nonlinearities. Guo and Zhu [18] established a sharp radial threshold separating scattering and finite-time blow-up for the L2L^2-supercritical fractional Hartree equation. Feng, Chen and Liu [14] obtained nonradial blow-up criteria and strong instability results for the fractional Schrödinger–Choquard equation. Earlier concentration and blow-up-profile results for mass-critical fractional Hartree equations were obtained, among others, by Cho–Hwang–Kwon–Lee [5].

Of particular relevance to the mass-critical problem, Lan [24] proved finite-time blow-up and obtained a detailed description of the blow-up dynamics for the one-dimensional L2L^2-critical fractional Schrödinger equation when the fractional dispersion exponent is sufficiently close to the classical Schrödinger case. More recently, Gou–Rădulescu–Zhang [21] extended localized-virial blow-up criteria for fractional NLS from radial to cylindrically symmetric data. Their mass-critical result, however, still retains the finite-time blow-up/infinite-time grow-up alternative. Thus, for the pure-power problem in dimensions N≥2N\ge2, the global grow-up branch in the general radial negative-energy mass-critical theory remains unresolved.

The purpose of this paper is to rule out this grow-up scenario for radial negative-energy solutions. Our main result is the following.

Theorem 1.1 (Finite-time blow-up for strong radial solutions). Assume (2). Let u∈C((−T−,T+);Hrad2s(RN))u \in C((-T_{-},T_{+});H_{\mathrm{rad}}^{2s}(\mathbb{R}^{N})) be a maximal strong solution of (1). If E(u0)<0E(u_{0}) < 0, then

T+<∞,T−<∞.T_{+} < \infty,\qquad T_{-} < \infty.

The H2sH^{2s} regularity in Theorem 1.1 is used only to justify the localized virial identity at the strong-solution level. All estimates obtained from the virial identity involve quantities at the HsH^{s} level. By establishing the integrated virial identity for HsH^{s} solution, we obtain the following energy-space formulation.

Corollary 1.2 (Energy-space formulation). Assume

N≥2,N2N−1≤s<1.(4)N \ge2,\qquad\frac{N}{2N-1} \le s < 1. \tag*{(4)}

Let u0∈Hrads(RN)u_{0} \in H_{\mathrm{rad}}^{s}(\mathbb{R}^{N}), and let u∈C((−T−,T+);Hrads(RN))u \in C((-T_{-},T_{+});H_{\mathrm{rad}}^{s}(\mathbb{R}^{N})) be the maximal solution furnished by the corresponding HsH^{s} local well-posedness theory. If E(u0)<0E(u_{0}) < 0, then

T+<∞,T−<∞.T_{+} < \infty,\qquad T_{-} < \infty.

In particular, the HsH^{s} norm diverges at each finite endpoint of the maximal lifespan.

The restriction N≥2N \ge2 in Theorem 1.1 enters through the radial Sobolev estimate used to control the nonlinear localization error, which requires

12<s<N2.\frac{1}{2} < s < \frac{N}{2}.

The additional lower bound

s≥N2N−1s \ge\frac{N}{2N-1}

in Corollary 1.2 comes from the currently available lossless radial local theory rather than from the virial argument itself; see [20, 8].

As a consequence, the known strong instability of the mass-critical ground-state standing waves can be strengthened to strong instability by genuine finite-time blow-up.

Corollary 1.3 (Strong instability by finite-time blow-up). Assume (4). Let Qω∈Hrads(RN)Q_{\omega} \in H_{\mathrm{rad}}^{s}(\mathbb{R}^{N}) be a positive ground state satisfying

(−Δ)sQω+ωQω−Qω1+4sN=0,ω>0.(5)(-\Delta)^{s}Q_{\omega}+\omega Q_{\omega}-Q_{\omega}^{1+\frac{4s}{N}}=0,\qquad\omega>0. \tag*{(5)}

Then, for every ε>0\varepsilon> 0, there exists u0∈Hrads(RN)u_{0} \in H_{\mathrm{rad}}^{s}(\mathbb{R}^{N}) such that

∥u0−Qω∥Hs<ε\lVert u_{0}-Q_{\omega}\rVert_{H^{s}} < \varepsilon

and the corresponding solution of (1) blows up in finite time.

Let us describe the main ideas behind Theorem 1.1.

A finite-time mechanism that has recently emerged in several mass-critical dispersive problems is the use of an exponentially localized virial field. In [25], the authors introduced such a multiplier for the four-dimensional mass-critical quadratic Schrödinger system without mass resonance. The resulting weighted kinetic defect, combined with a weighted interpolation estimate and a superlinear differential inequality, allows one to exclude the global grow-up scenario. Leonard Dinh [12] subsequently employed a related exponential localization for the homogeneous mass-critical biharmonic NLS and proved finite-time blow-up for radial negative-energy solutions, thereby eliminating the mass-critical grow-up alternative left open in the earlier biharmonic theory of Boulenger and Lenzmann [3]. More recently, in [26] the authors developed an analogous mechanism for the mass-critical inhomogeneous biharmonic NLS and obtained nonradial finite-time blow-up.

The essential new difficulty in the present problem is the nonlocality of the fractional dispersive operator. For second- and fourth-order local operators, the defect generated by a spatial multiplier can be expressed directly through local derivatives of the solution. For (−Δ)s(-\Delta)^s, this is no longer possible. The Balakrishnan representation rewrites the fractional kinetic energy in terms of the resolvent family

uμ=cs(−Δ+μ)−1u,u_\mu= c_s(-\Delta+\mu)^{-1}u,

and the localized fractional virial identity naturally produces a weighted resolvent quantity rather than a directly localized H˙s\dot{H}^s norm. Since the resolvent is itself nonlocal, multiplication by a spatial cutoff does not commute with it, and different dyadic regions continue to interact.

The mass-critical nonlinearity creates an additional difficulty. In the localized virial argument of Boulenger–Himmelsbach–Lenzmann [2], the localization function ϕR\phi_R agrees with the quadratic function on a large ball. Consequently,

I−D2ϕR=0,I-D^2\phi_R=0,

throughout this region, so the localization does not provide any control of the kinetic energy there. This is not a serious obstacle in the mass-supercritical case, where the localized virial identity contains a favorable coercive term even without such control. At the mass-critical exponent, however, the leading kinetic and nonlinear contributions cancel exactly. As a result, the localized virial estimate still shows that the virial observable decreases, but it does not provide sufficient control of the observable in terms of the remaining kinetic contribution. This is why the usual argument yields a blow-up-or-grow-up alternative rather than ruling out global solutions altogether.

Motivated by the exponential-localization mechanism above, we introduce

∇ϕR(x)=xexp⁡[−(∣x∣R)2s],\nabla\phi_R(x)=x\exp\left[-\left(\frac{|x|}{R}\right)^{2s}\right],

and

ωR(x)=1−exp⁡[−(∣x∣R)2s].\omega_R(x)=1-\exp\left[-\left(\frac{|x|}{R}\right)^{2s}\right].

The fractional virial identity naturally controls the Balakrishnan defect

QR(u):=∫0∞μs∫RNωR(x)∣∇uμ(x)∣2 dx dμ.Q_R(u):=\int_0^\infty\mu^s\int_{\mathbb{R}^N}\omega_R(x)|\nabla u_\mu(x)|^2\,dx\,d\mu.

To describe its localized Sobolev counterpart, let ρj=2jR\rho_j=2^jR, choose radial annular cutoffs ζj,ηj\zeta_j,\eta_j such that

∑j∈Zζj2=1,ηj≡1  on supp⁡ζj,\sum_{j\in\mathbb{Z}}\zeta_j^2=1,\qquad\eta_j\equiv1\ \text{ on }\operatorname{supp}\zeta_j,

and set

dj:=min⁡{(ρjR)2s,1}.d_j:=\min\left\{\left(\frac{\rho_j}{R}\right)^{2s},1\right\}.

Formally, the dyadic fractional defect is

DR(u):=∑j∈Zdj(∥(−Δ)s/2(ζju)∥L22+∥(−Δ)s/2(ηju)∥L22).D_R(u):=\sum_{j\in\mathbb{Z}}d_j\left(\|(-\Delta)^{s/2}(\zeta_j u)\|_{L^2}^2+\|(-\Delta)^{s/2}(\eta_j u)\|_{L^2}^2\right).

The precise construction of the cutoff families is given in § 4.

The first main step is the Balakrishnan-to-dyadic transfer estimate

DR(u)≲QR(u)+R−2sM(u).(6)D_R(u) \lesssim Q_R(u) + R^{-2s}M(u). \tag*{(6)}

It is proved by combining a dyadic decomposition with off-diagonal resolvent decay between separated annuli. This transfer is the principal new ingredient of the present argument: it converts the nonlocal coercivity furnished by the fractional virial identity into localized fractional Sobolev control.

The second step is to control the nonlinear localization error. Radial Sobolev estimates on the dyadic annuli yield, for every ε>0\varepsilon> 0,

∫RNωR(x)∣u(x)∣q dx≤εDR(u)+CεR−2sM(u)NN−2s.(7)\int_{\mathbb{R}^N} \omega_R(x)|u(x)|^q\,dx \le\varepsilon D_R(u) + C_\varepsilon R^{-2s}M(u)^{\frac{N}{N-2s}}. \tag*{(7)}

Combining (6) and (7) with the fractional virial identity gives the coercive localized virial estimate

MR′(t)≤8sE(u0)−cDR(u(t))+CR−2s[M(u0)+M(u0)NN−2s],\mathcal{M}_R'(t) \le8sE(u_0) - cD_R(u(t)) + CR^{-2s}\left[M(u_0) + M(u_0)^{\frac{N}{N-2s}}\right],

where

MR(u):=2Im⁡∫RNu‾∇ϕR⋅∇u dx.\mathcal{M}_R(u) := 2\operatorname{Im}\int_{\mathbb{R}^N} \overline{u}\nabla\phi_R \cdot\nabla u\,dx.

If E(u0)<0E(u_0) < 0, we may fix RR sufficiently large so that

References

  1. [1]Alampallam V. Balakrishnan, Fractional powers of closed operators and the semigroups generated by them, Pacific J. Math. 10 (1960), 419–437.
  2. [2]Thomas Boulenger, Dominik Himmelsbach, and Enno Lenzmann, Blowup for fractional NLS, J. Funct. Anal. 271 (2016), 2569–2603.arxiv.org/abs/1509.08845
  3. [3]Thomas Boulenger and Enno Lenzmann, Blowup for biharmonic NLS, Ann. Sci. Ec. Norm. Supér. (4) 50 (2017), no. 3, 503–544.DOI
  4. [4]Yonggeun Cho and Tohru Ozawa, Sobolev inequalities with symmetry, Commun. Contemp. Math. 11 (2009), 355–365.DOI
  5. [5]Yonggeun Cho, Gyeongha Hwang, Soonsik Kwon, and Sanghyuk Lee, Profile decompositions and blowup phenomena of mass critical fractional Schrödinger equations, Nonlinear Anal. 86 (2013), 12–29.arxiv.org/abs/1208.2303
  6. [6]Edward Brian Davies, Heat kernel bounds, conservation of probability and the Feller property, J. Analyse Math. 58 (1992), 99–119.DOI
  7. [7]Van Duong Dinh, Well-posedness of nonlinear fractional Schrödinger and wave equations in Sobolev spaces, Int. J. Appl. Math. 31 (2018), 483–525.
  8. [8]Van Duong Dinh, On blow-up solutions to the focusing mass-critical nonlinear fractional Schrödinger equation, Commun. Pure Appl. Anal. 18 (2019), 689–708.DOI
  9. [9]Van Duong Dinh, Blow-up criteria for fractional nonlinear Schrödinger equation, Nonlinear Anal. Real World Appl. 48 (2019), 117–140.DOI
  10. [10]Van Duong Dinh, On blowup solutions to the focusing L²-supercritical nonlinear fractional Schrödinger equation, J. Math. Phys. 59 (2018), 071506.
  11. [11]Van Duong Dinh, On instability of standing waves for the mass-supercritical fractional nonlinear Schrödinger equation, Z. Angew. Math. Phys. 70 (2019), Paper No. 58.
  12. [12]Leonard Dinh, Finite-time blow-up of radial negative-energy solutions for the mass-critical biharmonic nonlinear Schrödinger equation, online publication, (2026)
  13. [13]Binhua Feng, On the blow-up solutions for the fractional nonlinear Schrödinger equation with combined power-type nonlinearities, Comm. Pure Appl. Anal. 17 (2018), 1785–1804.
  14. [14]Binhua Feng, Ruipeng Chen, and Jiayin Liu, Blow-up criteria and instability of normalized standing waves for the fractional Schrödinger–Choquard equation, Adv. Nonlinear Anal. 10 (2021), 311–330.DOI
  15. [15]Rupert L. Frank, Enno Lenzmann, and Luis Silvestre, Uniqueness of radial solutions for the fractional Laplacian, Comm. Pure Appl. Math. 69 (2016), 1671–1726.
  16. [16]Rupert L. Frank and Robert Seiringer, Non-linear ground state representations and sharp Hardy inequalities, J. Funct. Anal. 255 (2008), 3407–3430.arxiv.org/abs/0803.0503
  17. [17]Robert T. Glassey, On the blowing up of solutions to the Cauchy problem for nonlinear Schrödinger equations, J. Math. Phys. 18 (1977), 1794–1797.DOI
  18. [18]Qing Guo and Shihui Zhu, Sharp threshold of blow-up and scattering for the fractional Hartree equation, J. Differential Equations 264 (2018), 2802–2832.DOI
  19. [19]Zihua Guo, Yannick Sire, Yuzhao Wang, and Lifeng Zhao, On the energy-critical fractional Schrödinger equation in the radial case, Dyn. Partial Differ. Equ. 15 (2018), no. 4, 265–282.DOI
  20. [20]Zihua Guo and Yuzhao Wang, Improved Strichartz estimates for a class of dispersive equations in the radial case and their applications to nonlinear Schrödinger and wave equations, J. Anal. Math. 124 (2014), 1–38.
  21. [21]Tianxiang Gou, Vicențiu D. Rădulescu, and Zhitao Zhang, Blow-up of cylindrically symmetric solutions for fractional NLS, Proc. Roy. Soc. Edinburgh Sect. A, published online 2025, doi:10.1017/prm.2025.10071.DOI
  22. [22]Younghun Hong and Yannick Sire, On fractional Schrödinger equations in Sobolev spaces, Commun. Pure Appl. Anal. 14 (2015), no. 6, 2265–2282.arxiv.org/abs/1501.01414
  23. [23]Nick Laskin, Fractional Schrödinger equation, Phys. Rev. E 66 (2002), 056108.pubmed.ncbi.nlm.nih.gov/12513557
  24. [24]Yang Lan, Blow-up dynamics for L²-critical fractional Schrödinger equations, Int. Math. Res. Not. IMRN 2022 (2022), no. 18, 13753–13810.DOI
  25. [25]Ngoc Uyen Cong Nguyen and Van Duong Dinh, Finite-time blow-up for the four-dimensional mass-critical quadratic nonlinear Schrödinger system without mass resonance, arXiv:2608.18453, 2026.arxiv.org/abs/2608.18453
  26. [26]Ngoc Uyen Cong Nguyen and Van Duong Dinh, Nonradial finite-time blow-up for the mass-critical inhomogeneous biharmonic NLS, arXiv:2610.24144, 2026.
  27. [27]Congming Peng and Qihong Shi, Stability of standing wave for the fractional nonlinear Schrödinger equation, J. Math. Phys. 59 (2018), 011508.DOI
  28. [28]Chenmin Sun, Hua Wang, Xiaohua Yao, and Jiqiang Zheng, Scattering below ground state of focusing fractional nonlinear Schrödinger equation with radial data, Discrete Contin. Dyn. Syst. 38 (2018), 2207–2228.
  29. [29]Takayoshi Ogawa and Yoshio Tsutsumi, Blow-up of H¹ solutions for the nonlinear Schrödinger equation, J. Differential Equations 92 (1991), 317–330.DOI

Paper details

vlnc:W_hhhn9e6d
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

GPT-5.6 Sol (bare)

View statement
Looking for related papers…
Contents