High-order asymptotic expansion for the nonlinear Klein-Gordon equation in the non-relativistic limit regime

Haohao Zheng; Jia Shen; Yanni Wang

REVIEW 2 major objections 4 minor 26 references

High-order asymptotic expansion for the nonlinear Klein-Gordon equation in the non-relativistic limit regime

T0 review · 2 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read The paper proves the first analytic high-order asymptotic expansion for the cubic nonlinear Klein-Gordon equation in the non-relativistic limit, with an $L^2$ error of order $\varepsilon^4+(\varepsilon^2t)^{\alpha/4}$ for $H^\alpha$…

desk verdict First analytic high-order expansion for NLKG non-relativistic limit with sharp rates, but the non-regular range α∈[4,8) has a genuine missing step in §5.2.3. read the letter →

arxiv 2411.13132 v1 pith:ORSHRIRJ submitted 2024-11-20 math.AP

Jia Shen , Yanni Wang , Haohao Zheng This is my paper · ORCID

classification math.AP MSC 35B4035B65

keywords nonlinearKlein-Gordonequationnon-relativisticlimitasymptoticexpansionSchrödingerprofilesconvergencerateStrichartzestimatesmodulatedFouriersharpness

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

This paper proves the first analytic high-order asymptotic expansion for the cubic nonlinear Klein-Gordon equation in the non-relativistic limit. Previous analytic results controlled the error of the leading Schrödinger profile by $\varepsilon^2+(\varepsilon^2t)^{\alpha/4}$ for $H^\alpha$ data with $\alpha\in[1,4]$; here a second profile is added and the leftover error is shown to be $\varepsilon^4+(\varepsilon^2t)^{\alpha/4}$ for $H^\alpha$ data with $\alpha\in[4,8]$. The correction consists of the third harmonic $\tfrac18 e^{3it/\varepsilon^2}v^3$ together with $e^{it/\varepsilon^2}w$, where $v$ solves the cubic nonlinear Schrödinger equation and $w$ solves an explicit Schrödinger-type equation chosen to cancel all order-$\varepsilon^2$ terms in the error equation. Counterexamples are constructed showing that this rate is optimal, up to a logarithmic factor for less regular data.

What carries the argument

The engine is the modulated Fourier (WKB) expansion: write the solution as a sum of terms $e^{imt/\varepsilon^2}$ times profiles with $\varepsilon$-independent time derivatives, and choose the profiles so that resonant terms cancel. The paper records three general lemmas: a forcing term of size $\varepsilon^a O(1)$, an oscillatory forcing $\varepsilon^{a-2}e^{imt/\varepsilon^2}G(t,x)$ with $m\neq1$, and a bootstrap-type nonlinearity each produce an order-$\varepsilon^a$ remainder. Applying these to the first-order error equation, the authors split the known order-$\varepsilon^2$ remainder into $\varepsilon^2\Phi_2+r_1$ and choose $\Phi_2$ so that all order-$\varepsilon^2$ and resonant terms vanish. The remaining error equation is controlled with Strichartz estimates, the fractional Leibniz rule, scattering bounds for the cubic nonlinear Schrödinger equation, a scaling transformation that removes $\varepsilon$ from the equation, and a bootstrap on the high-order remainder; a high-low frequency decomposition of the initial data reduces the regularity from $H^8$ to $H^\alpha$ with $\alpha\in[4,8]$.

What would settle it

Take the Gaussian initial data $g(x)=A_0^{d/2}e^{-A_0^2|x|^2}$ used in Section 6.3.1 and compute the $L^2$ difference between the exact nonlinear Klein-Gordon solution and the two-term profile for small $\varepsilon$ and $1\le t\le\varepsilon^{-2}\delta_0$; the paper's Lemma 6.1 predicts a lower bound growing like $\varepsilon^4t^2\|g\|_{\dot H^8}$. A direct numerical or analytic evaluation showing the error stays smaller than $c\varepsilon^4t^2$, or grows with a different power of $t$, would refute the sharpness claim.

Watch

Extended reading notes

Core claim

The central result is Theorem 1.2: for $d=2,3$ and $u_0,u_1\in H^\alpha(\mathbb{R}^d)$, $\alpha\in[4,8]$, the solution $u^\varepsilon$ of the nonlinear Klein-Gordon equation satisfies $\|u^\varepsilon-\Phi_1-\varepsilon^2\Phi_2\|_{L^2_x}\le C(\varepsilon^4+(\varepsilon^2t)^{\alpha/4})$ for all $t\ge0$. Here $\Phi_1=e^{it/\varepsilon^2}v+c.c.$ is the usual first-order Schrödinger profile and $\Phi_2=\tfrac18 e^{3it/\varepsilon^2}v^3+e^{it/\varepsilon^2}w+c.c.$ is the new second-order profile, with $v$ solving $2i\partial_t v-\Delta v+3|v|^2v=0$ and $w$ solving the Schrödinger-type equation (1.3) whose initial value is fixed by matching the initial conditions of the remainder. The proof isolates the order-$\varepsilon^2$ part of the first-order error, absorbs it into $\Phi_2$, and shows the remaining profile $r_1$ satisfies an equation whose nonlinearities all fall into classes known to produce order-$\varepsilon^4$ errors. Theorem 1.5 supplies matching lower bounds, so the $\varepsilon^4$ term and the $(\varepsilon^2t)^{\alpha/4}$ growth are optimal.

Load-bearing premise

The whole proof leans on a previously established first-order estimate: the leading error is at most $\varepsilon^2$ plus $\varepsilon^4T$ times a fixed Sobolev norm of the initial data, under the smallness condition that $\varepsilon^4T$ times that norm is below a fixed constant. If that estimate fails or holds only on shorter times, the new $\varepsilon^4$ convergence rate and its $(\varepsilon^2t)^{\alpha/4}$ growth would have to be modified.

Editorial extensions

If this is right

For $H^8$ data the error is $O(\varepsilon^4(1+t+t^2))$: two terms of the expansion are enough to push the small-$\varepsilon$ error from order $\varepsilon^2$ to order $\varepsilon^4$.
For less regular data the error bound degrades as $(\varepsilon^2t)^{\alpha/4}$, so the expansion remains quantitatively useful over times up to roughly $\varepsilon^{-2}\delta_0$ before the time factor dominates.
The third harmonic $\tfrac18 e^{3it/\varepsilon^2}v^3$ is forced by the cubic nonlinearity; any correct second-order expansion must contain it, up to the stated freedom of absorbing terms into $w$.
The counterexamples show the bound $\varepsilon^4+(\varepsilon^2t)^{\alpha/4}$ cannot be improved, up to a logarithmic factor for $H^\alpha$ data.
The choice of $w$ is not unique, but all admissible decompositions lead to the same convergence rate.

Reading between the lines

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

Inference: the same cancellation mechanism should extend to still higher orders, with each additional $\varepsilon^2$ level adding odd harmonics $e^{(2j+1)it/\varepsilon^2}$ and requiring roughly four more derivatives of the data, following the regularity ladder $H^{4k}$ for a $k$-th order expansion.
Inference: the lower-bound analysis isolates the operator $1+\tfrac{i}{8}t\Delta^2$ as the dominant transport of the remainder, suggesting that the next correction beyond the nonlinear Schrödinger dynamics is a fourth-order (biharmonic) term; this could be tested by comparing the Klein-Gordon solution with the NLS evolution plus a biharmonic correction.
Inference: numerical integrators built on this two-term profile should show $\varepsilon^4$ accuracy for $H^8$ data, and the sharp time growth gives a concrete benchmark for when the second correction is necessary.

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

Short version: this is the first analytic high-order expansion in this regime, and the formal part is genuinely good. But the proof of Theorem 1.2 for α<8 has a gap in Section 5.2.3 that I would want fixed before trusting the stated range.

What's new: Lei-Wu gave the first-order error ε²+(ε²t)^{α/4}. The authors add the ε²Φ₂ profile (the explicit ⅛e^{3it/ε²}v³ plus a corrector w), prove the error is ε⁴+(ε²t)^{α/4} for H^α data with α∈[4,8], and give lower-bound counterexamples showing the rates are sharp. If true, this is the right next step and it explains the numerical observations of Schratz-Zhao and Faou-Schratz.

What's good: the derivation of Φ₂ via modulated Fourier expansion is clear, and the equation for w is natural. The bootstrap structure with Lemmas 3.1–3.3 is a sensible extension of Lei-Wu. The proof is long but mostly structured: Strichartz estimates for the scaled KG equation, scattering theory for cubic NLS, and a two-regime bootstrap. The counterexamples in Section 6 are serious work; the lower-bound construction using e^{it(-⟨∇⟩+1-Δ/2)}=1+⅛itΔ²+... is a believable route. I did not find a fatal flaw in the regular case (H^8 data).

The soft spot: Section 5.2.3, where the paper reduces from H^8 to H^α. The authors define v_N and w_N with low-frequency initial data and then claim ||uε-Φ₁N-ε²Φ₂N|| is handled by 'a similar argument to Proposition 5.1'. But Proposition 5.1 applies to the original uε with profiles matching the full initial data. Here uε is the original solution, so uε-Φ₁N-ε²Φ₂N has initial data mismatch from the high-frequency part P>N u₀, and the nonlinearity contains Φ₁-Φ₁N times the remainder. That residual forcing is never written down, and the bootstrap in Lemmas 3.1–3.3 would require controlling terms like (Φ₁-Φ₁N)·r₁ in the S_sc/Z_sc norms. The N-balances in Subcases A1–B3 do not display these terms; they only balance the profile differences and the smooth-profile error. It is plausible the gap is fixable by introducing an auxiliary solution with filtered initial data and adding a separate estimate for the difference, but that argument is absent. For α=8 the issue is harmless because the regular case applies, but for α∈[4,8) this is an omitted step in the main theorem.

There is also a smaller point: the smallness condition (5.21) uses ˙H^{4+sc}, but (5.22)–(5.23) are then stated with N-powers without spelling out how the bootstrap interval size and the eventual ε²t terms interact. This is cosmetic compared with the gap above.

Bottom line: this is a worthwhile paper with a novel result and a mostly credible proof, but the non-regular case needs repair. It deserves a serious referee; I would send it out, asking the authors to close the Section 5.2.3 gap or to state clearly that Theorem 1.2 is proved only for regular data.

Referee Report

2 major / 4 minor

Summary. The paper studies the cubic nonlinear Klein-Gordon equation (1.1) in the non-relativistic limit regime for d=2,3 and proves an analytic high-order asymptotic expansion. The proposed expansion is u_ε ≈ Φ_1 + ε^2 Φ_2, where Φ_1 = e^{it/ε^2}v + c.c., Φ_2 = (1/8)e^{3it/ε^2}v^3 + e^{it/ε^2}w + c.c., v solves the cubic NLS (1.2), and w solves the Schrödinger-type equation (1.3). Theorem 1.2 claims the L^2 error is O(ε^4 + (ε^2t)^{α/4}) for H^α initial data with α∈[4,8]. Theorem 1.5 gives matching lower bounds by counterexamples. The proof strategy is to derive a residual equation for r_1, use Strichartz estimates and bootstrap arguments in scaled variables, prove the H^8 regular case, and then reduce to lower regularity by a high-low frequency decomposition of the initial data.

Significance. If the main theorem is correct, this is the first analytic second-order asymptotic expansion for this non-relativistic limit, improving the first-order result of Lei and Wu from ε^2 + (ε^2t)^{α/4} with α∈[1,4] to ε^4 + (ε^2t)^{α/4} with α∈[4,8]. The expansion coefficients are not fitted: they are determined by exact cancellation of the O(ε^2) residuals, and the lower bounds in Theorem 1.5 are constructed independently, so the claimed rate is genuinely falsifiable. The paper makes heavy but clearly identified use of external inputs: global scattering for the cubic NLS and the Lei–Wu first-order estimate. The overall architecture of the proof is credible, but two load-bearing points need correction before the main claim is fully established.

major comments (2)

[§5.2.3] The reduction from Theorem 1.2 for α<8 to the regular case is incomplete. After defining Φ_1N and Φ_2N in (5.18), the proof bounds ‖u_ε−Φ_1N−ε^2Φ_2N‖ by 'a similar argument to the proof of Proposition 5.1'. Proposition 5.1, however, applies to a remainder with zero initial data and with profiles generated by the same initial data as the NLKG solution. Here u_ε has the full initial data (u_0,u_1), while the profiles are generated by P_≤N v_0. The difference u_ε(0)−Φ_1N(0)−ε^2Φ_2N(0) contains the P_>N part of the initial data, of size N^{−α}, and the time derivative has a similar P_>N contribution of size ε^{−2}N^{−α}. The equation, initial data, and forcing terms for w := u_ε−Φ_1N−ε^2Φ_2N are never written down, so the bootstrap lemmas of §3 are not demonstrably available for this function; in particular, the nonlinearity contains differences such as |Φ_1|^2Φ_1−|Φ_1N|^2Φ_1N multiplied by w, and the displayed N-balances in Subcases A1–B3 never verify the smallness of such products in the Z_sc and S_sc norms. This is not a notational issue: for α=4 with the chosen N=(ε^2T)^{−1/4}, the initial mismatch is N^{−α}=(ε^2T)^{α/4}, which is exactly the target rate and cannot be discarded without an argument.
[§5.1 and Lemma 4.2] There is an inconsistency in the time normalization between Proposition 5.1 and Lemma 4.2. Lemma 4.2 is stated for the scaled variable R_1 satisfying the ε-independent equation (4.6), with a smallness condition ε^4T‖v_0‖_{Ḣ^{4+sc}}≤δ0. The final scaling formula (5.4) maps R_1 on [0,ε^{−2}T] back to r_1 on [0,T]. Therefore, when the proof of Proposition 5.1 is carried out for an original time interval of length T, the correct smallness condition is ε^2T‖v_0‖_{Ḣ^{4+sc}}≤δ0, not ε^4T‖v_0‖_{Ḣ^{4+sc}}≤δ0. The regular-case proof nevertheless splits the time axis using ε^4T, while the analogous non-regular condition (5.21) correctly uses ε^2T. As written, Lemma 4.2 is applied on a longer interval than its smallness condition allows. The threshold and the original/scaled interval notation should be made consistent throughout §5.1.

minor comments (4)

[§5.2.1] The symbol v_N is used both for the low-frequency solution of (5.6) and for the high-frequency difference v−v_N in (5.7); this makes (5.8)–(5.12) ambiguous. A separate notation such as v^N for the difference would clarify the estimates.
[§4.2.2, equation (4.16)] The symbol ∂ttth appears to be a typo: the displayed expansion describes the third time derivative of h, so it should be ∂ttth or ∂_t^3h.
[References] References [15] and [16] appear to cite the same Faou–Schratz paper with identical title and journal data; if so, one duplicate should be removed.
[Abstract] The phrase 'There are extensive numerical and analytic results concerning that the solution...' should read 'concerning the fact that the solution...' or 'showing that...'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the profiles are fixed by residual cancellation, the cited first-order estimate is external, and the counterexamples independently extract the dominant Duhamel term.

full rationale

The derivation chain contains no step in which a claimed prediction reduces to an input by construction. The high-order profile Φ2 in Definition 1.1 is not fitted to the solution uε: the coefficient (1/8)e^{3it/ε²}v³ is forced by requiring the coefficient of e^{±3it/ε²} in the renormalized error equation to vanish (Section 3.2, with v³−8η2=0), and w is chosen to cancel the ε²e^{±it/ε²} coefficients; the initial data w0 are algebraically determined by matching r(0) and ∂tr(0). The remaining equation for r1 is then solved by Strichartz and bootstrap estimates, not assumed. The first-order estimate quoted in Lemma 4.2, ‖R‖_{S_sc} ≲ ε²+ε⁴T‖v0‖, is taken from Lei and Wu [21], an external paper with stated assumptions, and is used as a genuine input with its smallness condition; it is not a self-citation and not the conclusion being proved. The lower bounds in Theorem 1.5 are derived independently by identifying the dominant term I1(t) in the Duhamel formula and estimating it from below; they do not presuppose Theorem 1.2. One proof gap should be recorded: in Section 5.2.3 the estimate of uε−Φ1N−ε²Φ2N is dismissed with 'we apply a similar argument to the proof of Proposition 5.1', and the equation and initial data for this filtered remainder are never written. This is an omitted justification, not a circular step: nothing in the paper defines that remainder to equal the desired bound. There are no fitted parameters, no load-bearing self-citations, no uniqueness assertion imported from the authors, and no renaming of a known result as a new one. The central claim therefore has independent analytic content.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The expansion contains no fitted numbers: the profiles v and w are fixed by cancellation conditions and existing equations. The only external inputs are standard Strichartz and scattering results plus the cited first-order theorem. No new physical or mathematical entities are postulated.

assumptions (6)

standard math Strichartz estimates for linear Schrödinger and Klein-Gordon equations
Invoked repeatedly in Sections 3-6, recorded as Lemmas 2.3 and 2.4.
standard math Fractional Leibniz rule
Lemma 2.1, used for product and nonlinearity estimates.
domain assumption Global well-posedness and scattering for defocusing cubic NLS in 2D and 3D
Lemma 2.6, needed to obtain uniform space-time bounds on the Schrödinger profile v.
domain assumption Lei-Wu first-order remainder estimate
Estimate (4.39) from [21] is load-bearing in the bootstrap control of the high-order remainder R1.
standard math Mihlin-Hörmander multiplier theorem
Used in Section 6.2 to bound the operator Tt in the phase approximation.
domain assumption Initial data u0,u1∈H^α with α∈[4,8] and v0=1/2(u0-iu1)
This is the theorem's hypothesis; the regularity range determines the convergence rate.

how reviews work

0 comments

Cite this review

Pith. "Pith review of High-order asymptotic expansion for the nonlinear Klein-Gordon equation in the non-relativistic limit regime." pith.science (2026). https://pith.science/paper/ORSHRIRJ

@misc{pith2026241113132,
  author       = {Pith},
  title        = {Pith review of: High-order asymptotic expansion for the nonlinear Klein-Gordon equation in the non-relativistic limit regime},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ORSHRIRJ}},
  note         = {Machine review of arXiv:2411.13132}
}

abstract

This paper presents an investigation into the high-order asymptotic expansion for 2D and 3D cubic nonlinear Klein-Gordon equations in the non-relativistic limit regime. There are extensive numerical and analytic results concerning that the solution of NLKG can be approximated by first-order modulated Schr\"odinger profiles in terms of $e^{i\frac t {\varepsilon^2}}v + c.c. $, where $v$ is the solution of related NLS and ``$c.c.$" denotes the complex conjugate. Particularly, the best analytic result up to now is given in \cite{lei}, which proves that the $L_x^2$ norm of the error can be controlled by $\varepsilon^2 +(\varepsilon^2t)^{\frac \alpha 4}$ for $H^\alpha_x$-data, $\alpha \in [1, 4]$. As for the high-order expansion, to our best knowledge, there are only numerical results, while the theoretical one is lacking. In this paper, we extend this study further and give the first high-order analytic result. We introduce the high-order expansion inspired by the numerical experiments in \cite{schratz2020, faou2014a}: \[ e^{i\frac t {\varepsilon^2}}v +\varepsilon^2 \Big( \frac 18 e^{3i\frac t {\varepsilon^2} }v^3 +e^{i\frac t {\varepsilon^2}} w \Big) +c.c., \] where $w$ is the solution to some specific Schr\"odinger-type equation. We show that the $L_x^2$ estimate of the error is of higher order $\varepsilon^4+\left(\varepsilon^2t\right)^\frac{\alpha}{4}$ for $H^\alpha_x$-data, $\alpha \in [4, 8]$.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 25 canonical work pages

[1]

Bao and Y

W. Bao and Y. Cai. Uniform error estimates of ﬁnite diﬀere nce methods for the nonlinear Schr¨ odinger equation with wave operator. SIAM J. Numer. Anal. , 50(2):492–521, 2012. 1.1, 1.1

work page 2012
[2]

Bao and Y

W. Bao and Y. Cai. Uniform and optimal error estimates of a n exponential wave integrator sine pseudospectral method for the nonlinear Schr¨ odinger equation with wave op erator. SIAM J. Numer. Anal. , 52(3):1103–1127,

work page
[3]

W. Bao, Y. Cai, and X. Zhao. A uniformly accurate multisca le time integrator pseudospectral method for the Klein-Gordon equation in the nonrelativistic limit regime . SIAM J. Numer. Anal. , 52(5):2488–2511, 2014. 1.1

work page 2014
[4]

Bao and X

W. Bao and X. Dong. Analysis and comparison of numerical m ethods for the Klein-Gordon equation in the nonrelativistic limit regime. Numer. Math. , 120(2):189–229, 2012. 1.1

work page 2012
[5]

W. Bao, Y. Lu, and Z. Zhang. Convergence rates in the nonre lativistic limit of the cubic Klein-Gordon equation,

work page
[6]

Bao and X

W. Bao and X. Zhao. A uniformly accurate multiscale time i ntegrator spectral method for the Klein-Gordon- Zakharov system in the high-plasma-frequency limit regime . J. Comput. Phys. , 327:270–293, 2016. 1.1

work page 2016
[7]

Bao and X

W. Bao and X. Zhao. Comparison of numerical methods for th e nonlinear Klein-Gordon equation in the nonrel- ativistic limit regime. J. Comput. Phys. , 398:108886, 2019. 1.1

work page 2019
[8]

Chartier, N

P. Chartier, N. Crouseilles, M. Lemou, and F. M´ ehats. Un iformly accurate numerical schemes for highly oscilla- tory Klein-Gordon and nonlinear Schr¨ odinger equations. Numer. Math. , 129(2):211–250, 2015. 1.1

work page 2015

Show all 26 references

[9]

Chartier, J

P. Chartier, J. Makazaga, A. Murua, and G. Vilmart. Multi -revolution composition methods for highly oscillatory diﬀerential equations. Numer. Math. , 128(1):167–192, 2014. 1.1

2014
[10]

F. M. Christ and M. I. Weinstein. Dispersion of small amp litude solutions of the generalized Korteweg-de Vries equation. J. Funct. Anal. , 100(1):87–109, 1991. 2.2 38 JIA SHEN, YANNI W ANG, AND HAOHAO ZHENG

1991
[11]

Cohen, E

D. Cohen, E. Hairer, and C. Lubich. Modulated Fourier ex pansions of highly oscillatory diﬀerential equations. Found. Comput. Math. , 3(4):327–345, 2003. 1.1

2003
[12]

B. Dodson. Global well-posedness and scattering for th e defocusing, L2-critical, nonlinear Schr¨ odinger equation when D=2. Duke Math. J. , 165(18), 2016. 2.2

2016
[13]

X. Dong, Z. Xu, and X. Zhao. On time-splitting pseudospe ctral discretization for nonlinear Klein-Gordon equa- tion in nonrelativistic limit regime. Commun. Comput. Phys. , 16(2):440–466, 2014. 1.1

2014
[14]

E. Faou, L. Gauckler, and C. Lubich. Sobolev stability o f plane wave solutions to the cubic nonlinear Schr¨ odinger equation on a torus. Commun. Partial Diﬀer. Equ. , 38(7):1123–1140, 2013. 1.1

2013
[15]

Faou and K

E. Faou and K. Schratz. Asymptotic preserving schemes f or the Klein–Gordon equation in the non-relativistic limit regime. Numer. Math. , 126(3):441–469, 2014. (document), 1.1, 1.1

2014
[16]

Faou and K

E. Faou and K. Schratz. Asymptotic preserving schemes f or the Klein-Gordon equation in the non-relativistic limit regime. Numer. Math. , 126(3):441–469, 2014. 1.1

2014
[17]

Ginibre and G

J. Ginibre and G. Velo. Smoothing properties and retard ed estimates for some dispersive evolution equations. Commun. Math. Phys. , 144(1):163–188, 1992. 2.2

1992
[18]

Hairer, M

E. Hairer, M. Hochbruck, A. Iserles, and C. Lubich. Geom etric numerical integration. Oberwolfach Rep., 3(1):805– 882, 2006. 1.1

2006
[19]

M. A. Keel and T. Tao. Endpoint Strichartz estimates. Amer. J. Math. , 120(5):955–980, 1998. 2.2

1998
[20]

Killip, T

R. Killip, T. Tao, and M. Visan. The cubic nonlinear Schr ¨ odinger equation in two dimensions with radial data. J. Eur. Math. Soc. , 11(6):1203–1258, 2009. 2.2

2009
[21]

Lei and Y

Z. Lei and Y. Wu. Non-relativistic limit for the cubic no nlinear Klein-Gordon equations. arXiv preprint arXiv:2309.10235, 2023. (document), 1.1, 1.1, 1.2, 1.4, 1.2, 3.5, 3.2, 4.2.2

2023 arXiv
[22]

J. E. Lin and W. A. Strauss. Decay and scattering of solut ions of a nonlinear Schr¨ odinger equation.J. Functional Analysis, 30(2):245–263, 1978. 2.2

1978
[23]

Masmoudi and K

N. Masmoudi and K. Nakanishi. From nonlinear Klein-Gor don equation to a system of coupled nonlinear Schr¨ odinger equations.Math. Ann. , 324:359–389, 2002. 1.1

2002
[24]

Masmoudi and K

N. Masmoudi and K. Nakanishi. Nonrelativistic limit fr om Maxwell-Klein-Gordon and Maxwell-Dirac to Poisson- Schr¨ odinger.Int. Math. Res. Not. , (13):697–734, 2003. 1.1

2003
[25]

Schratz and X

K. Schratz and X. Zhao. On comparison of asymptotic expa nsion techniques for nonlinear Klein-Gordon equation in the nonrelativistic limit regime. Discrete Contin. Dyn. Syst.-Ser. B , 25(8):2841–2865, 2020. (document), 1.1, 1.1

2020
[26]

X. Zhao. A combination of multiscale time integrator an d two-scale formulation for the nonlinear Schr¨ odinger equation with wave operator. J. Comput. Appl. Math. , 326:320–336, 2017. 1.1 Jia Shen School of Mathematical Sciences and LPMC, Nankai Universit y, Tianjin 300071, C...

2017

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Bao and Y

W. Bao and Y. Cai. Uniform error estimates of ﬁnite diﬀere nce methods for the nonlinear Schr¨ odinger equation with wave operator. SIAM J. Numer. Anal. , 50(2):492–521, 2012. 1.1, 1.1

work page 2012

[2] [2]

Bao and Y

W. Bao and Y. Cai. Uniform and optimal error estimates of a n exponential wave integrator sine pseudospectral method for the nonlinear Schr¨ odinger equation with wave op erator. SIAM J. Numer. Anal. , 52(3):1103–1127,

work page

[3] [3]

W. Bao, Y. Cai, and X. Zhao. A uniformly accurate multisca le time integrator pseudospectral method for the Klein-Gordon equation in the nonrelativistic limit regime . SIAM J. Numer. Anal. , 52(5):2488–2511, 2014. 1.1

work page 2014

[4] [4]

Bao and X

W. Bao and X. Dong. Analysis and comparison of numerical m ethods for the Klein-Gordon equation in the nonrelativistic limit regime. Numer. Math. , 120(2):189–229, 2012. 1.1

work page 2012

[5] [5]

W. Bao, Y. Lu, and Z. Zhang. Convergence rates in the nonre lativistic limit of the cubic Klein-Gordon equation,

work page

[6] [6]

Bao and X

W. Bao and X. Zhao. A uniformly accurate multiscale time i ntegrator spectral method for the Klein-Gordon- Zakharov system in the high-plasma-frequency limit regime . J. Comput. Phys. , 327:270–293, 2016. 1.1

work page 2016

[7] [7]

Bao and X

W. Bao and X. Zhao. Comparison of numerical methods for th e nonlinear Klein-Gordon equation in the nonrel- ativistic limit regime. J. Comput. Phys. , 398:108886, 2019. 1.1

work page 2019

[8] [8]

Chartier, N

P. Chartier, N. Crouseilles, M. Lemou, and F. M´ ehats. Un iformly accurate numerical schemes for highly oscilla- tory Klein-Gordon and nonlinear Schr¨ odinger equations. Numer. Math. , 129(2):211–250, 2015. 1.1

work page 2015