Description and classification of 2-solitary waves for nonlinear damped Klein-Gordon equations

Lifeng Zhao; Rapha\"el C\^ote; Xu Yuan; Yvan Martel

Description and classification of 2-solitary waves for nonlinear damped Klein-Gordon equations

T0 review · 0 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read For the damped Klein-Gordon equation, any two-soliton wave must consist of opposite-sign solitons whose separation follows a universal law: |z1-z2| = log t - (N-1)/2 log log t + c0 + o(1).

desk verdict A serious, technically complete classification of damped Klein-Gordon 2-solitary waves; the proof is careful and the soft spots are minor — send it to a strong referee. read the letter →

arxiv 1908.09527 v1 pith:JUKUOTPD submitted 2019-08-26 math.AP

Raphaël Côte , Yvan Martel , Xu Yuan , Lifeng Zhao This is my paper · ORCID

classification math.AP MSC 35L7135B4037K40

keywords dampedKlein-Gordonequation2-solitarywavesgroundstatelinearizedoperatorlogarithmicdistanceunstablemanifoldLipschitzgraphclassificationsolitonresolution

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper studies two-soliton waves of the nonlinear damped Klein-Gordon equation $\partial_{tt}u+2\alpha\partial_t u-\Delta u+u-|u|^{p-1}u=0$ in dimensions $1\le N\le 5$ with energy-subcritical $p>2$. It establishes a complete dichotomy: two solitary waves with the same sign cannot persist as a two-soliton wave, while a pair with opposite signs must separate in a universal way, with distance asymptotic to $\log t-\frac{N-1}{2}\log\log t+c_0$. The initial data that produce such opposite-sign two-soliton waves form, near the sum of two remote solitons, a codimension-2 Lipschitz graph: only the unstable component around each soliton is free, and it is uniquely determined by the separation and the stable remainder. A sympathetic reader would care because this gives a fully parameter-free description of a non-integrable multi-soliton interaction, showing that damping turns repulsive soliton pairs into a deterministic logarithmic law.

What carries the argument

The load-bearing object is the linearized operator $L=-\Delta+1-pQ^{p-1}$ around the unique ground state $Q$. The paper uses Lemma 1.7: $L$ has a unique negative eigenvalue $-\nu_0^2$ with normalized eigenfunction $Y$, and a coercivity bound separating $\langle\varepsilon,Y\rangle$ and the translation modes $\langle\varepsilon,\partial_{x_j}Q\rangle$. Around this, the proof builds a modulation decomposition with geometric parameters $z_k,\ell_k$ and orthogonality conditions, sharp estimates for the interaction term $G=f(Q_1+Q_2)-f(Q_1)-f(Q_2)$, including the sign-dependent asymptotics $\langle G,\nabla Q_1\rangle\approx \sigma c_1 \frac{z}{|z|} g(|z|)$, a modified energy $E$ defined in (2.34), and a Lyapunov functional $M$. These yield the trichotomy of Section 2.4: the distance obeys $\frac{d}{dt}[1/q(r)]\approx -\sigma g_0/\alpha$, the unstable components satisfy $\dot b\approx 2\nu_+ b$, and the damped components decay. The sign $\sigma=\sigma_1\sigma_2$ in the distance ODE is what creates the dichotomy: $\sigma=-1$ integrates to $\log t$, while $\sigma=+1$ has no global solution.

What would settle it

Find, numerically, a same-sign two-soliton wave: run the damped Klein-Gordon equation from initial data close to $Q(x-L/2)+Q(x+L/2)$ with small unstable components and check whether the solution remains a two-soliton wave for large time. The theorems imply instead that the energy drops below $2E(Q,0)$ and such a configuration cannot persist; a robust computed counterexample would falsify Theorem 1.3. For the universal law, measure $z_1-z_2$ for an opposite-sign pair and compare the limit of $|z_1-z_2|-(\log t-\frac{N-1}{2}\log\log t)$ with the predicted constant $\log(\kappa g_0/\alpha)$; a different constant or a different rate would falsify Theorem 1.4.

Watch

Extended reading notes

Core claim

The paper's central discovery is that two-soliton waves of the damped Klein-Gordon equation are completely described by the sign of each soliton and by one unstable amplitude per soliton. Theorem 1.3 rules out equal signs. For opposite signs, Theorem 1.4 gives a decomposition $u(t)=\sigma_1 Q(\cdot-z_1(t))+\sigma_2 Q(\cdot-z_2(t))+\varepsilon(t)$ with $\|\varepsilon\|_{H^1}+\|\partial_t u\|_{L^2}\lesssim t^{-1}$, a universal distance asymptotics (1.8) with dimension-dependent constant $c_0(N)$, and a fixed separation direction $\omega_\infty$ as in (1.9). Theorem 1.5 then proves that, for initial data within $\delta$ of $Q(\cdot-L/2)-Q(\cdot+L/2)$ with $|L|>10|\log\delta|$, the solution is a two-soliton wave exactly when the two unstable components $(h_1,h_2)$ lie on a Lipschitz graph $H(L,\varphi)$, with $|H|\lesssim e^{-L/2}+\|\varphi\|$. In short: damping plus interaction forces a universal repulsion, and the only free data are the unstable modes.

Load-bearing premise

The argument rests on the spectral coercivity of the linearized operator $L$ around one ground state: a unique negative eigenvalue with eigenfunction $Y$, and the bound $\langle L\varepsilon,\varepsilon\rangle \ge c\|\varepsilon\|_{H^1}^2 - c^{-1}(\langle\varepsilon,Y\rangle^2 + \sum_j \langle\varepsilon,\partial_{x_j}Q\rangle^2)$; if this failed, the one-dimensional unstable direction could not be separated from the stable remainder.

Editorial extensions

If this is right

Same-sign two-soliton waves do not exist: any candidate configuration must either break apart or lose one soliton before settling.
Every opposite-sign two-soliton wave has centers whose difference satisfies the universal asymptotic (1.8) with a dimension-dependent constant; the direction of separation is fixed.
The residual error decays like $t^{-1}$, so the two-soliton wave is asymptotically a sum of two translated ground states with algebraic accuracy.
Near the sum of two remote opposite-sign solitons, the two-soliton waves form a codimension-2 Lipschitz manifold: $h=H(L,\varphi)$ with $H$ Lipschitz and of size at most $C(e^{-L/2}+\|\varphi\|)$.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The same distance ODE suggests that multi-soliton waves with three or more bubbles would have pairwise separations all growing like $\log t$, with alternating signs; this is not proved in the paper.
The leading $\log t$ rate is independent of the damping coefficient $\alpha$, but the constant $c_0$ depends on $\alpha$ through $\log(\kappa g_0/\alpha)$; measuring that constant for two values of $\alpha$ would be a direct quantitative test.
The nonexistence of same-sign pairs relies on the energy decreasing to $2E(Q,0)$; this mechanism may transfer to other damped semilinear wave equations whose linearized operator has a single unstable mode, but the paper does not treat those equations.
A numerical experiment initializing two opposite-sign solitons with small unstable components and checking whether $|z_1-z_2|-(\log t-\frac{N-1}{2}\log\log t)$ tends to a constant would test the universality claim directly.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Two things you should know. First, this paper really does what the title says: it describes and classifies 2-solitary waves for the damped nonlinear Klein-Gordon equation, with nonexistence for same signs, universal log t separation for opposite signs, and a codimension-2 Lipschitz classification of initial data. Second, the proof is long but careful, in the standard Côte–Martel modulation/bootstrap style, and on a second pass I did not find a load-bearing hole.

The genuinely new content is the universal log t distance, which is specific to the damping and contrasts with the linear-in-time rates in undamped models. Equally new is the full classification of the family, showing that only the unstable directions in the initial data matter. The nonexistence of same-sign waves is new and the energy argument behind it is clean. The construction and uniqueness parts (Propositions 4.2 and 4.4) are technically demanding but coherent; the retraction argument is standard and correctly set up.

Soft spots are minor. The spectral/coercivity input (Lemma 1.7) is a black box from the literature, but it is exactly the standard property of the linearized Klein-Gordon operator, so that is acceptable. More substantive: Remark 4.3 asserts that earlier multi-soliton constructions in [6,7,21] contain a gap related to modulation, and the present paper works around it. That claim is plausible and the paper's own construction is clean, but a referee should independently verify it; a published correction of earlier work deserves scrutiny. Finally, the constant c0 in Theorem 1.4 is stated as c0 = c0(N), but the proof gives c0 = log(κg0/α), which depends on α and p. Since α and p are fixed throughout, this is cosmetic, but the statement should match the proof.

Bottom line: this paper is for analysts working on multi-soliton dynamics or damped wave equations, and it will likely become a reference for the damped Klein-Gordon case. It deserves a serious referee, and my own verdict is accept, with requests to fix the c0 notation and to add a bit more detail on the correction of [6,7,21] if the editor wants to preempt controversy.

Referee Report

0 major / 4 minor

Summary. The paper gives a complete dynamical description of 2-solitary waves for the damped nonlinear Klein-Gordon equation (1.1) in dimensions 1 <= N <= 5 and energy-subcritical p > 2. It proves three main results: Theorem 1.3 establishes nonexistence of same-sign 2-solitary waves; Theorem 1.4 shows that every opposite-sign 2-solitary wave has a universal asymptotic distance between the centers, |z1(t) - z2(t)| = log t - ((N-1)/2) log log t + c0 + o(1), together with convergence of the direction of separation; Theorem 1.5 classifies all 2-solitary waves near the sum of two remote solitons as a codimension-2 Lipschitz graph H(L, phi) = h. The proof combines modulation theory, refined nonlinear interaction estimates (Lemma 2.1), a modified energy with coercivity (Lemma 2.4), a bootstrap trichotomy for the distance, the unstable direction, and the damped components (Lemma 2.6 and Proposition 3.1), and a topological no-retraction argument with a contraction-mapping uniqueness step in Section 4.

Significance. If the results hold, the paper constitutes a substantial contribution to the theory of multi-solitary waves: it gives the first complete description for a damped dispersive equation, including a universal logarithmic separation law that is qualitatively different from the linear-in-time separation in undamped problems. The classification result is sharp in the sense that the stable/unstable structure of each soliton leads to exactly a codimension-2 family. The proofs are detailed and essentially self-contained except for standard well-posedness and spectral inputs, which are explicitly cited. A notable strength is the explicit correction, in Remark 4.3, of a technical flaw in earlier multi-soliton constructions in [6,7,21], and the use of exactly modulated initial data to avoid it. The leading-order constant c0 is computed explicitly rather than introduced as a free parameter, and the asymptotic law (1.8)-(1.9) is falsifiable by numerical simulation. I found no load-bearing error in the modulation, bootstrap, energy, or topological arguments.

minor comments (4)

[Theorem 1.4, Eq. (1.8) and proof in Section 3.3] The theorem states c0 = c0(N), but the proof derives c0 = log(kappa g0 / alpha), which depends on the damping parameter alpha as well as on N. Since alpha is fixed throughout the paper this does not affect the central claims, but the statement should be corrected to c0 = c0(N, alpha) or to an explicit formula.
[Proposition 3.1, estimate (3.3)] The constant ~C in (3.3) is said to depend on R±(T_delta), T_delta, and delta, but the displayed estimate also has a term t/|log delta|; it would help the reader to state explicitly that the implied constant is independent of t as t -> infinity, or to separate the transient and asymptotic parts.
[Reference list, item [28]] The reference is to V. E. Zakharov and A. B. Shabat; the first author's initial is incorrectly printed as 'T. Zakharov'.
[Lemma 2.2(iv) and Lemma 2.6] The exponent theta is introduced as any 1 < theta < min(p-1,2) in several places; for readability, the identical range could be defined once at the beginning of Section 2, since the repeated statements are easy to miss.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the 2-solitary wave description is derived from modulation/energy estimates rather than assumed.

full rationale

The paper's central claims are derived, not assumed. The universal log t asymptotic in Theorem 1.4 follows from Proposition 3.2, which uses the ODE for 1/q(r) obtained in Lemma 2.6 (estimates (2.46), (2.52)) together with the bootstrap and damping estimates; the asymptotic is not put into the definition of a 2-solitary wave. Theorem 1.3 uses the energy expansion (2.54) and the monotonicity of the damped energy (1.3), so the same-sign nonexistence is a consequence of an energy comparison rather than a restatement of the definition. Theorem 1.5 is proved by constructing the codimension-2 manifold through Lemma 4.9, a contraction argument in Proposition 4.10, and the uniqueness/Lipschitz estimate of Proposition 4.4, which compares two actual 2-solitary waves; no fitted parameter is relabeled as a prediction. The spectral and coercivity input of Lemma 1.7 is cited to [7, Lemma 1], but it is a standard external property of the linearized operator L, parameter-free and not containing the target results, so it does not make the argument circular. The paper even explicitly identifies and corrects a flaw in earlier self-cited constructions in Remark 4.3, which shows the self-citations are used as background rather than as load-bearing justification. The only notable issue is notational: Theorem 1.4 writes c0 = c0(N), while the proof yields c0 = log(κg0/α), which depends on the fixed damping parameter α; this is a presentation issue, not a circularity. Overall, the derivation chain is self-contained relative to standard elliptic and Cauchy theory, and no step reduces by construction to its own inputs.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The paper introduces no ad hoc fitted parameters or new entities. The constants c0, g0, and κ are determined from the ground state and the equation, not fitted. All axioms are standard results quoted from the literature.

assumptions (3)

standard math The ground state Q exists, is unique, positive, radial, and satisfies the decay estimate q(r) ~ κ r^{-(N-1)/2} e^{-r} as r→∞ (equation (1.11)).
Invoked throughout for interaction estimates, e.g., Lemma 2.1. This is a known result from [2,17], not proved in the paper.
domain assumption Local well-posedness of the Cauchy problem in H^1 × L^2 (from [3, Theorem 2.3]).
Used to define solutions and ensure the energy identity (1.3). It is a standard result for this damped equation.
standard math Spectral property of L: unique negative eigenvalue -ν0^2 with normalized eigenfunction Y, and the coercivity estimate of Lemma 1.7.
Central to the modulation decomposition and energy estimates (Lemmas 2.2, 2.4). It is cited from [7, Lemma 1].

how reviews work

0 comments

Cite this review

Pith. "Pith review of Description and classification of 2-solitary waves for nonlinear damped Klein-Gordon equations." pith.science (2026). https://pith.science/paper/JUKUOTPD

@misc{pith2026190809527,
  author       = {Pith},
  title        = {Pith review of: Description and classification of 2-solitary waves for nonlinear damped Klein-Gordon equations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JUKUOTPD}},
  note         = {Machine review of arXiv:1908.09527}
}

abstract

We describe completely 2-solitary waves related to the ground state of the nonlinear damped Klein-Gordon equation \begin{equation*} \partial_{tt}u+2\alpha\partial_{t}u-\Delta u+u-|u|^{p-1}u=0 \end{equation*} on $\bf R^N$, for $1\leq N\leq 5$ and energy subcritical exponents $p>2$. The description is twofold. First, we prove that 2-solitary waves with same sign do not exist. Second, we construct and classify the full family of 2-solitary waves in the case of opposite signs. Close to the sum of two remote solitary waves, it turns out that only the components of the initial data in the unstable direction of each ground state are relevant in the large time asymptotic behavior of the solution. In particular, we show that $2$-solitary waves have a universal behavior: the distance between the solitary waves is asymptotic to $\log t$ as $t\to \infty$. This behavior is due to damping of the initial data combined with strong interactions between the solitary waves.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

28 extracted references · 27 canonical work pages

[1]

Bates and C.K.R.T

P.W. Bates and C.K.R.T. Jones, Invariant manifolds for semilinear partial diﬀerential eq ua- tions. Dynamics reported, Vol. 2, 1–38, Dynam. Report. Ser. Dynam. Systems Appl., 2, Wiley, Chichester, 1989

work page 1989
[2]

Berestycki and P.-L

H. Berestycki and P.-L. Lions, Nonlinear scalar ﬁeld equ ations. I. Existence of a ground state. Arch. Rational Mech. Anal. 82, (1983) 313–345

work page 1983
[3]

N. Burq, G. Raugel and W. Schlag, Long time dynamics for da mped Klein-Gordon equations. Ann. Sci. ´Ec. Norm. Sup´ er.(4) 50 (2017), no. 6, 1447–1498

work page 2017
[4]

Combet, Multi-soliton solutions for the supercritic al gKdV equations

V. Combet, Multi-soliton solutions for the supercritic al gKdV equations. Comm. Partial Diﬀerential Equations 36 (2011), no. 3, 380–419

work page 2011
[5]

Combet and Y

V. Combet and Y. Martel, Construction of multi-bubble so lutions for the critical gKdV equation. SIAM J. Math. Anal. , 50(4) (2018), 3715–3790

work page 2018
[6]

Cˆ ote, Y

R. Cˆ ote, Y. Martel and F. Merle, Construction of multi-s oliton solutions for the L2- supercritical gKdV and NLS equations. Rev. Mat. Iberoamericana 27 (2011), 273–302

work page 2011
[7]

Cˆ ote and C

R. Cˆ ote and C. Mu˜ noz, Multi-solitons for nonlinear Klein-Gordon equations. Forum of Math- ematics, Sigma. 2 (2014). 34 R. C ˆOTE, Y. MARTEL, X. YUAN, AND L. ZHAO

work page 2014
[8]

Duyckaerts, C

T. Duyckaerts, C. E. Kenig and F. Merle, Classiﬁcation of radial solutions of the focusing, energy-critical wave equation. Cambridge J. Math. , 1 (2013), 75–144

work page 2013

Show all 28 references

[9]

Duyckaerts, H

T. Duyckaerts, H. Jia, C. E. Kenig and F. Merle, Soliton re solution along a sequence of times for the focusing energy critical wave equation. Geom. Funct. Anal., 27 (2017), no. 4, 798–862

2017
[10]

Feireisl, Convergence to an equilibrium of semiline ar wave equations on unbounded inter- vals

E. Feireisl, Convergence to an equilibrium of semiline ar wave equations on unbounded inter- vals. Dynam.Systems Appl. 3 (1994), 423–434

1994
[11]

Feireisl, Finite energy travelling waves for nonlin ear damped wave equations

E. Feireisl, Finite energy travelling waves for nonlin ear damped wave equations. Quart. Appl. Math. 56 (1998), no. 1, 55–70

1998
[12]

M. A. Jendoubi, A remark on the convergence of global and bounded solutions for a semilinear wave equation on RN . J. Dynam. Diﬀerential Equations 14 (2002), no. 3, 589–596

2002
[13]

Jendrej, Nonexistence of radial two-bubbles with op posite signs for the energy-critical wave equation Ann

J. Jendrej, Nonexistence of radial two-bubbles with op posite signs for the energy-critical wave equation Ann. Sc. Norm. Super. Pisa Cl. Sci. (5) Vol. XVIII (2018), 1–44

2018
[14]

Jendrej, Construction of two-bubble solutions for e nergy-critical wave equations

J. Jendrej, Construction of two-bubble solutions for e nergy-critical wave equations. Amer. J. Math. 141 (2019), no. 1, 55–118

2019
[15]

Jendrej and A

J. Jendrej and A. Lawrie, Two-bubble dynamics for thres hold solutions to the wave maps equation. Invent. Math. 213 (2018), no. 3, 1249–1325

2018
[16]

Jendrej, Dynamics of strongly interacting unstable two-solitons for generalized Korteweg-de Vries equations

J. Jendrej, Dynamics of strongly interacting unstable two-solitons for generalized Korteweg-de Vries equations. Preprint arXiv:1802.06294

arXiv
[17]

M. K. Kwong, Uniqueness of positive solutions of ∆ u − u + up = 0 in Rn. Arch. Rational Mech. Anal. 105 (1989), no. 3, 243–266

1989
[18]

Krieger and W

J. Krieger and W. Schlag, Stable manifolds for all monic supercritical focusing nonlinear Schr¨ odinger equations in one dimension. J. Amer. Math. Soc. 19 (2006), 815–920

2006
[19]

Krieger, K

J. Krieger, K. Nakanishi and W. Schlag, Global dynamics above the ground state energy for the one-dimensional NLKG equation. Math. Z. 272 (2012), no. 1-2, 297–316

2012
[20]

Li and L

Z. Li and L. Zhao, Asymptotic decomposition for nonline ar damped Klein-Gordon equations, Preprint arXiv:1511.00437

arXiv
[21]

Martel and F

Y. Martel and F. Merle, Construction of multi-solitons for the energy-critical wave equation in dimension 5. Arch. Ration. Mech. Anal. 222 (2016), no. 3, 1113–1160

2016
[22]

Martel, F

Y. Martel, F. Merle, K. Nakanishi and P. Rapha¨ el, Codim ension one threshold manifold for the critical gKdV equation, Comm. Math. Phys. , 342 (2016), 1075–1106

2016
[23]

Martel and T

Y. Martel and T. V. Nguyen, Construction of 2-solitons w ith logarithmic distance for the one-dimensional cubic Schr¨ odinger system. Preprint arXiv:1903.07175

1903 arXiv
[24]

R. M. Miura, The Korteweg-de Vries equation, a survey of results. SIAM Review 18 (1976), 412–459

1976
[25]

Nakanishi and W

K. Nakanishi and W. Schlag, Invariant manifolds and dispersive Hamiltonian evolution equa- tions. Z¨ urich Lectures in Advanced Mathematics. European Mathem atical Society (EMS), Z¨ urich, 2011

2011
[26]

T. V. Nguyen, Strongly interacting multi-solitons wit h logarithmic relative distance for the gKdV equation. Nonlinearity, 30(12):4614, 2017

2017
[27]

T. V. Nguyen, Existence of multi-solitary waves with lo garithmic relative distances for the NLS equation. C. R. Math. Acad. Sci. Paris 357 (2019), no. 1, 13–58

2019
[28]

Zakharov and A.B

T. Zakharov and A.B. Shabat. Exact theory of two-dimens ional self-focusing and one- dimensional self-modulation of waves in nonlinear media. Sov. Phys. JETP 34 (1972), 62–69. IRMA UMR 7501, Universit ´e de Strasbourg, CNRS, F-67000 Strasbourg, France E-mail address : cote@mat...

1972

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Bates and C.K.R.T

P.W. Bates and C.K.R.T. Jones, Invariant manifolds for semilinear partial diﬀerential eq ua- tions. Dynamics reported, Vol. 2, 1–38, Dynam. Report. Ser. Dynam. Systems Appl., 2, Wiley, Chichester, 1989

work page 1989

[2] [2]

Berestycki and P.-L

H. Berestycki and P.-L. Lions, Nonlinear scalar ﬁeld equ ations. I. Existence of a ground state. Arch. Rational Mech. Anal. 82, (1983) 313–345

work page 1983

[3] [3]

N. Burq, G. Raugel and W. Schlag, Long time dynamics for da mped Klein-Gordon equations. Ann. Sci. ´Ec. Norm. Sup´ er.(4) 50 (2017), no. 6, 1447–1498

work page 2017

[4] [4]

Combet, Multi-soliton solutions for the supercritic al gKdV equations

V. Combet, Multi-soliton solutions for the supercritic al gKdV equations. Comm. Partial Diﬀerential Equations 36 (2011), no. 3, 380–419

work page 2011

[5] [5]

Combet and Y

V. Combet and Y. Martel, Construction of multi-bubble so lutions for the critical gKdV equation. SIAM J. Math. Anal. , 50(4) (2018), 3715–3790

work page 2018

[6] [6]

Cˆ ote, Y

R. Cˆ ote, Y. Martel and F. Merle, Construction of multi-s oliton solutions for the L2- supercritical gKdV and NLS equations. Rev. Mat. Iberoamericana 27 (2011), 273–302

work page 2011

[7] [7]

Cˆ ote and C

R. Cˆ ote and C. Mu˜ noz, Multi-solitons for nonlinear Klein-Gordon equations. Forum of Math- ematics, Sigma. 2 (2014). 34 R. C ˆOTE, Y. MARTEL, X. YUAN, AND L. ZHAO

work page 2014

[8] [8]

Duyckaerts, C

T. Duyckaerts, C. E. Kenig and F. Merle, Classiﬁcation of radial solutions of the focusing, energy-critical wave equation. Cambridge J. Math. , 1 (2013), 75–144

work page 2013