Pith. sign in

REVIEW 4 major objections 7 minor 53 references

An exponential-free Runge--Kutta framework for developing third-order unconditionally energy stable schemes for the Cahn--Hilliard equation

T0 review · 4 major / 7 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 Taylor replacements for exponentials give Cahn–Hilliard time steppers that conserve mass, preserve equilibria, and never raise the free energy.

desk verdict Good scheme, honest numerics; the two main proofs don't close as written, but both are fixable. read the letter →

arxiv 2411.16271 v1 pith:57344Z2C submitted 2024-11-25 math.NA cs.NA

Haifeng Wang , Jingwei Sun , Hong Zhang , Xu Qian , Songhe Song This is my paper · ORCID
classification math.NAcs.NA MSC 65M0665L0635K5565M70
keywords Cahn–Hilliardequationexponential-freeRunge–Kuttaunconditionalenergystabilityequilibrium-statepreservationFourierspectralcollocationconvergenceanalysisadaptivetimestepping
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 develops a family of time-stepping methods for the Cahn–Hilliard equation that keep the original free energy from increasing no matter how large the time step. The starting point is Lawson's integrating-factor Runge–Kutta method, whose matrix exponentials damp high Fourier modes and therefore smear interfacial layers and pull steady states away from their true equilibria. The authors replace those exponentials by Taylor polynomials, obtaining exponential-free Runge–Kutta (EFRK) schemes. They prove that the resulting first-, second-, and third-order schemes conserve mass, decrease the discrete Ginzburg–Landau energy for every $\tau>0$ when the stabilization parameter is large enough, and satisfy an $\ell^2$ error estimate $C(h^m+\tau^p)$. If correct, the schemes make long-time coarsening simulations with adaptive time steps reliable without nonlinear solvers.

What carries the argument

The central object is the Taylor-polynomial matrix function $\phi_k(z)=1+z+\cdots+z^k/k!$ used in place of $e^z$ in Lawson's integrating-factor Runge–Kutta step: each stage inverts $\phi_i(-c_i\tau L_\kappa)$ instead of applying $e^{-c_i\tau L_\kappa}$. This choice makes the equilibrium-preservation condition $\hat d_i\equiv I$ hold exactly for the first three tableaux. The argument is carried by Lemma 4.1 and Lemma 4.2, which rewrite $L_\kappa u_{n,i}+N_\kappa(u_{n,i-1})$ as a linear combination of consecutive stage differences, and by the energy-stability matrices of Table 2 whose positive semi-definiteness turns the energy difference into a sum of non-positive quadratic forms. The stabilization operator $L_\kappa=\Delta_N(-\epsilon^2\Delta_N+\kappa)$ and the parameter $\kappa$ keep the scheme linear in the unknown and explicit in the nonlinear term.

What would settle it

Run EFRK(3,3) with the original cubic double-well potential, $\kappa=2$, and a time step $\tau$ larger than the stability threshold, starting from a random field in $[-0.5,0.5]^2$; monitor $\max\|u_{n,i}\|_{\ell^\infty}$ and the discrete energy at every stage. If any stage leaves $|u|>\sqrt{15}/3$ while the energy rises over that step, then the unconditional-stability theorem as stated is violated for the original nonlinearity.

Watch

Extended reading notes

Core claim

The central claim is that Taylor-polynomial approximations of the exponential, rather than the exponentials themselves, are enough to get high-order, unconditionally energy-stable integration of the Cahn–Hilliard equation. Theorem 4.4 states that if the stabilization parameter satisfies $\kappa \ge \max_{|\xi|\le \beta}|f'(\xi)|/2$ with $\beta=\max\|u_{n,i}\|_{\ell^\infty}$ and if the energy-stability matrices in Table 2 are positive semi-definite, then $E(u^{n+1})\le E(u^n)$ for any $\tau>0$. The paper verifies these matrices for the RK(1,1), Heun RK(2,2), and Heun RK(3,3) tableaux in (10), gives an affirmative answer to equilibrium preservation via the condition $\hat d_i\equiv I$, and proves mass conservation and the error estimate. The experiments show EFRK(3,3) matching reference solutions under adaptive stepping where IFRK(3,3) drifts.

Load-bearing premise

The proof assumes the numerical solution stays inside a band $|u|\le\beta$ so that $\kappa \ge \max_{|\xi|\le\beta}|f'(\xi)|/2$; the Cahn–Hilliard equation has no maximum principle, and the paper does not prove such a bound for the original cubic potential, so the unconditional-stability statement relies on that unverified (or truncated-potential) condition.

Editorial extensions

If this is right

  • Large or adaptive time steps become usable for long-time coarsening simulations: for any $\tau>0$ the discrete energy $E(u^n)$ does not increase, so runs to $T=100$ with $\tau=10^{-2}$ remain stable.
  • Equilibrium states are fixed points of the time integrator, so a steady interface profile is not smoothed or drifted by exponential damping; numerical solutions stay at local minima rather than climbing away from them.
  • The $\ell^2$ error satisfies $\|I_h u(t_n)-u^n\|_{\ell^2} \le C(h^m+\tau^p)$ for $p=1,2,3$, so third-order EFRK gives accuracy comparable to reference solutions at much larger steps.
  • The unified energy-stability matrix framework gives a recipe to check unconditional energy stability for any explicit Runge–Kutta tableau whose coefficients satisfy $\hat d_i\equiv I$; in particular first- and second-order EFRK are also unconditionally stable.
  • The same construction extends to other gradient-flow phase-field models such as epitaxial growth and phase-field crystal equations, as the paper states in its conclusion.

Reading between the lines

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

  • Editorial extension: because the proof only needs a Taylor-polynomial replacement and the sign of $\Delta_N^{-1}$, the equilibrium-preservation argument should transfer to any $H^{-1}$ gradient flow whose spatial operator is a symmetric negative semi-definite matrix; this is not a claim the paper makes.
  • The paper shows that no four-stage fourth-order explicit Runge–Kutta tableau can satisfy the equilibrium-preservation constraints, so a natural testable extension is whether a fifth-order or non-explicit tableau, or a higher-degree Taylor polynomial, can restore fourth order while keeping unconditional stability.
  • The unproved $\ell^\infty$ bound could be checked computationally: track $\max\|u_{n,i}\|_{\ell^\infty}$ during adaptive runs; if it exceeds $\beta=\sqrt{15}/3$ yet energy still decreases, one could try to prove a weaker bound or reduce the stabilization error.
  • The linear-stability result that $\theta=1/2$ gives A-stability suggests that the optimal stabilization parameter equals half the Lipschitz constant of the nonlinearity; future work could test whether $\kappa=(3\beta^2-1)/2$ is robust across interface thicknesses.
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.

Referee Report

4 major / 7 minor

Summary. The paper introduces a class of exponential-free Runge--Kutta (EFRK) schemes for the Cahn--Hilliard equation, obtained by replacing the exponential evolution operators of Lawson-type integrating factor Runge--Kutta methods with Taylor polynomials. The authors show that this replacement preserves nontrivial equilibrium states, conserves mass, and yields first- to third-order time-stepping schemes. They further claim unconditional energy stability for the proposed stabilization EFRK methods, provide a linear stability analysis, and derive an ℓ2 error estimate of the form C(h^m + τ^p). The paper concludes with numerical experiments, including long-time coarsening simulations and an adaptive time-stepping strategy.

Significance. If the theoretical claims were fully established, the EFRK framework would be a useful addition to the toolkit of high-order, unconditionally energy-stable schemes for phase-field models, especially because of its equilibrium-preservation property and the unified energy-stability proof based on 'energy stability matrices.' The paper also provides reproducible code, which is commendable. However, two load-bearing theoretical results are not supported as stated: the unconditional energy stability proof depends on an a posteriori ℓ∞ bound that is not proven for the original cubic nonlinearity, and the error estimate has a constant that grows exponentially with the number of time steps, so it does not imply convergence as τ → 0. These issues currently limit the significance of the paper's central claims, although the numerical experiments and the underlying framework remain of interest.

major comments (4)
  1. [§4.1, Lemma 4.3 and Remark 4.3] The energy-stability proof is circular because the stabilization assumption is formulated in terms of β = max_{n,i} ||u_{n,i}||_{ℓ∞}, which is defined from the computed EFRK stages themselves. Consequently the hypothesis κ ≥ max_{|ξ|≤β} |f'(ξ)|/2 cannot be verified a priori, and Theorem 4.4 does not establish unconditional energy stability 'for any τ > 0' for the original cubic potential. Remark 4.3 explicitly concedes that no maximum-norm bound is proven for the EFRK solution and suggests truncating the potential instead. As written, the theorem at best covers the modified potential (23); the abstract and conclusion overclaim preservation of the original energy dissipation without time-step restrictions.
  2. [§5, Theorem 5.3 and Eq. (42)–(44)] The error estimate is not a genuine convergence estimate because the constant C2 = C5 sqrt(2^{n+1} t_n C3^2 e^{C4 s t_n}) depends on n, the number of time steps. With t_n = nτ, the factor 2^{n+1} behaves like 2^{T/τ} for fixed final time T, so the right-hand side of (44) does not tend to zero as τ → 0 (in fact it diverges). The transition from the recursion (41) to the bound (42) hides this exponential-in-1/τ growth, and the claimed ℓ2 error estimate C(h^m + τ^p) in Theorem 5.4 and in the abstract is therefore unsupported.
  3. [§5, Theorem 5.3, discussion after Eq. (34)] The proof states 'Acting I − c_i τ Lκ on both sides of (34)' when the equation involves φ_i(−c_i τ Lκ), which equals I − c_i τ Lκ only for i = 1; for i = 2,3 there are additional positive semidefinite terms. The displayed inequality (35) can be salvaged because φ_i(−c_i τ Lκ) ≥ I − c_i τ Lκ in the positive semidefinite sense, but this step is not explained and should be made explicit.
  4. [§4.2, Linear stability analysis] The claim that EFRK(3,3) is A-stable precisely at θ = 1/2 is supported only by a figure and the phrase 'through careful calculation.' Since the paper presents this as a theoretical result, a proof or a detailed verification of |Φ(θ,z)| ≤ 1 on the left half-plane should be provided; otherwise the statement should be weakened to a numerical observation.
minor comments (7)
  1. [§1, p. 2] In the historical overview, 'proved the inherit the properties' should read 'proved that it inherits the properties.'
  2. [§2, p. 4] The definition of the grid function space writes 'M_N1 = {v|...} ∈ R^{N1}'; the symbol should be ⊂ or ≅, not ∈.
  3. [§3.1, Eq. (8)] The underbrace annotation '=0' beneath part of the expression is ambiguous and should be reworded for clarity.
  4. [§4.1, Lemma 4.3] The phrase 'the two adjacent stage solutions, as derived from...' is grammatically awkward; consider 'For any two adjacent stage solutions u_{n,i-1}, u_{n,i} obtained from...'
  5. [§4.1, Table 2] Table 2 is garbled in the manuscript; the entries for the energy stability matrices need to be typeset clearly, as they are central to the proof of Theorem 4.4.
  6. [§5, Theorem 5.3] The regularity assumption writes 'H^p(0,T; H^{m+d+4}_{pre}(Ω))'; the subscript should be 'per' for periodic.
  7. [§4.2 and §5] The phrase 'A-stable for any time step' is imprecise: A-stability is a property of the stability function, not of the time step. Also, the constants C3 and C4 in the proof of Theorem 5.3 are defined inline but would benefit from being labeled consistently in the display.

Circularity Check

2 steps flagged · score 6.0 of 10

The 'unconditional' energy-stability theorem is conditional on an unproven, solution-dependent ℓ∞ bound in Lemma 4.3; the central claim for the original Cahn–Hilliard nonlinearity is not self-contained.

  1. self definitional [Lemma 4.3, Theorem 4.4 (Section 4.1)]
    "Assume that the stabilization parameter satisfies κ ≥ max_{|ξ|≤β} 1/2 |f'(ξ)|, where β = max_{n,i}{∥u_{n,i}∥_{ℓ∞}} ... E(un,i) − E(un,i−1) ≤ ⟨un,i − un,i−1, ∆^{-1}_N [Lκun,i + Nκ(un,i−1)]⟩"

    The hypothesis of Lemma 4.3 defines β as the maximum ℓ∞ norm of the computed EFRK stages. Consequently the energy-dissipation inequality and Theorem 4.4's conclusion 'unconditionally energy stable for any τ > 0' hold only for numerical solutions that are already known to be bounded by β. That boundedness is exactly what the stability analysis would need to establish; it is instead assumed through the definition of β. The proof therefore reduces the stability claim to a property of the very solution it is supposed to certify. The paper concedes in Remark 4.3 that 'it has not been proven that the solution obtained by EFRK is bounded in maximum norm,' so the theorem cannot verify the hypothesis a priori for the original cubic nonlinearity f(u)=u^3−u.

  2. other [Remark 4.3, Eq. (23)]
    "the Cahn–Hilliard equation does not inherently possess a maximum bound principle, and it has not been proven that the solution obtained by EFRK is bounded in maximum norm. Therefore, a common practice [10] is to modify the potential by truncating it when |u| > β ... the potential is replaced with a quadratic function ... Consequently, the nonlinear term f(u) is replaced by f̃(u)"

    To make the solution-dependent assumption of Lemma 4.3 satisfiable, the paper changes the problem itself: the cubic double-well potential is replaced by a truncated quadratic extension with |f̃′(u)| ≤ 3β²−1. The unconditional energy-stability analysis then applies to this modified nonlinearity, not to the Cahn–Hilliard equation with f(u)=u³−u as stated in the abstract and introduction. The claim that the schemes 'preserve the original energy dissipation without time-step restrictions' therefore relies on either an unproved bound for the original equation or on silently substituting a different equation. The numerical choice κ=2 is likewise calibrated from the observed empirical bound β=√15/3, so the demonstration does not independently certify the original problem.

full rationale

The construction of the EFRK schemes themselves is not circular: the Taylor-polynomial coefficients and Butcher tableaux are fixed by order conditions and the equilibrium-preservation requirements d_i≡I, and the energy-stability matrices are worked out explicitly for the first- to third-order schemes. Mass conservation follows from symmetry of the matrix functions, linear stability is an independent calculation based on the stability function Φ(θ,z), and the numerical experiments are benchmarked against independent ERK(3,3) reference solutions. No fitted parameter is being renamed as a prediction, and the cited prior work by the same group ([34], [45]) is used as a source of a construction technique and an error bound rather than as a substitute for the main stability derivation. However, the central theorem advertised as 'unconditionally energy stable for any τ>0' for the original Cahn–Hilliard equation is materially weakened by Lemma 4.3's hypothesis β=max_{n,i}∥u_{n,i}∥_∞, which is a property of the computed solution. Because this bound is unproved and is explicitly acknowledged as unproved in Remark 4.3, the stability proof for the original cubic nonlinearity is conditional in a way that depends on the object of the analysis; for the truncated potential the claim is valid but concerns a modified equation. Overall circularity score 6: the central 'prediction' of unconditional energy dissipation for the original problem reduces, at its key step, to an assumption defined in terms of the numerical solution.

Assumptions & free parameters 2 free parameters · 6 assumptions · 0 invented entities

The central claim rests on standard spectral approximation and matrix-function theory plus the usual regularity assumptions for parabolic error estimates. The nonstandard burden is the ell-infinity bound or potential truncation needed to select kappa; no new physical entities are introduced.

free parameters (2)
  • stabilization coefficient kappa = kappa = 2 in numerical experiments; theory requires kappa >= max_{|xi| <= beta} |f'(xi)|/2
    Introduced in Eq. (5) as L_kappa = Delta_N(-epsilon^2 Delta_N + kappa) to stabilize the explicit treatment of nonlinearity. All energy-stability and convergence results depend on its size; the experiments set kappa = 2 from the heuristic bound beta = sqrt(15)/3 in Remark 4.3.
  • truncation level beta for double-well potential = beta = sqrt(15)/3 approximately 1.291
    In Remark 4.3 the potential is truncated at +/- beta to enforce the Lipschitz bound needed by Lemma 4.3. The choice of beta is justified by the observation that the phase variable is always bounded in maximum norm of order O(1) and by numerical tests, not by an a priori proof.
assumptions (6)
  • standard math Spectral interpolation error bound: ||partial_x^l(I_N v - v)||_{L2} <= C* h^{m-l} for v in H^m_per (Lemma 5.1).
    Used in Theorem 5.2 to bound the spatial discretization defect delta(t); standard spectral method result from Shen-Tang-Wang.
  • domain assumption Numerical solution remains bounded in ell-infinity (beta = max ||u_{n,i}||_infinity) or the double-well potential is truncated at +/- beta.
    Lemma 4.3 and Theorem 4.4 require kappa >= max_{|xi| <= beta} |f'(xi)|/2 over the range of the numerical solution. For the cubic f, this needs an unproven bound or a modified potential; Remark 4.3 admits this and uses beta = sqrt(15)/3 empirically.
  • domain assumption Exact solution has Sobolev regularity u in H^1(0,T; H^{m+d+4}_per) for the spatial estimate and u in H^p(0,T; H^{m+d+4}_per) for the temporal estimate.
    The error estimates in Theorems 5.2-5.4 assume enough smoothness of the exact solution; standard in numerical analysis of parabolic PDEs.
  • standard math Matrix functions phi_i(-c_i tau L_kappa) satisfy spectral mapping, symmetry, and commutativity properties (Lemma 3.1 from Higham).
    Used in the mass-conservation proof and in the energy-stability computations.
  • domain assumption Delta_N is self-adjoint and negative definite on the zero-mean subspace, so Delta_N^{-1} exists (Remark 2.1).
    Energy-stability estimates write terms as <., Delta_N^{-1}.>; valid for the zero-mean Fourier collocation discretization of periodic Cahn-Hilliard.
  • standard math Gronwall inequality and Sobolev embedding theorems used in Theorems 5.2 and 5.3.
    Standard analytic tools for deriving a priori error bounds.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An exponential-free Runge--Kutta framework for developing third-order unconditionally energy stable schemes for the Cahn--Hilliard equation." pith.science (2026). https://pith.science/paper/57344Z2C

@misc{pith2026241116271,
  author       = {Pith},
  title        = {Pith review of: An exponential-free Runge--Kutta framework for developing third-order unconditionally energy stable schemes for the Cahn--Hilliard equation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/57344Z2C}},
  note         = {Machine review of arXiv:2411.16271}
}
abstract

In this work, we develop a class of up to third-order energy-stable schemes for the Cahn--Hilliard equation. Building on Lawson's integrating factor Runge--Kutta method, which is widely used for stiff semilinear equations, we discuss its limitations, such as the inability to preserve the equilibrium state and the oversmoothing of interfacial layers in the solution's profile because of the exponential damping effects. To overcome this drawback, we approximate the exponential term using a class of sophisticated Taylor polynomials, leading to a novel Runge--Kutta framework called exponential-free Runge--Kutta. By incorporating stabilization techniques, we analyze the energy stability of the proposed schemes and demonstrate that they preserve the original energy dissipation without time-step restrictions. Furthermore, we perform an analysis of the linear stability and establish an error estimate in the $\ell^2$ norm. A series of numerical experiments validate the high-order accuracy, mass conservation, and energy dissipation of our schemes.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 47 canonical work pages

  1. [1]

    Free energy of a nonuniform system

    John W Cahn and John E Hilliard. Free energy of a nonuniform system. I. Interfacial free energy. The Journal of chemical physics, 28(2):258–267, 1958

    work page 1958

  2. [2]

    Models for phase separation and their mathematics

    Paul C Fife. Models for phase separation and their mathematics. Electronic Journal of Differential Equations , 2000

    work page 2000

  3. [3]

    A stable and conservative finite difference scheme for the Cahn-Hilliard equation

    Daisuke Furihata. A stable and conservative finite difference scheme for the Cahn-Hilliard equation. Numerische Math- ematik, 87(4):675–699, 2001

    work page 2001

  4. [4]

    An unconditionally stable one-step scheme for gradient systems

    David J Eyre. An unconditionally stable one-step scheme for gradient systems. Unpublished article, 6, 1998

    work page 1998

  5. [5]

    Unconditionally gradient stable time marching the Cahn-Hilliard equation

    David J Eyre. Unconditionally gradient stable time marching the Cahn-Hilliard equation. MRS online proceedings library (OPL), 529:39, 1998

    work page 1998

  6. [6]

    Unconditionally stable finite difference, nonlinear multigrid simulation of the Cahn-Hilliard-Hele-Shaw system of equations

    Steven M Wise. Unconditionally stable finite difference, nonlinear multigrid simulation of the Cahn-Hilliard-Hele-Shaw system of equations. Journal of Scientific Computing , 44(1):38–68, 2010

    work page 2010

  7. [7]

    A third order accurate in time, BDF-type energy stable scheme for the Cahn-Hilliard equation

    Kelong Cheng, Cheng Wang, Steven M Wise, and Yanmei Wu. A third order accurate in time, BDF-type energy stable scheme for the Cahn-Hilliard equation. Numer. Math. Theory Methods Appl. , 15(2):279–303, 2022

    work page 2022

  8. [8]

    An adaptive time-stepping strategy for the molecular beam epitaxy models

    Zhonghua Qiao, Zhengru Zhang, and Tao Tang. An adaptive time-stepping strategy for the molecular beam epitaxy models. SIAM Journal on Scientific Computing , 33(3):1395–1414, 2011

    work page 2011

Show all 53 references
  1. [9]

    An energy-stable and convergent finite-difference scheme for the phase field crystal equation

    Steven M Wise, Cheng Wang, and John S Lowengrub. An energy-stable and convergent finite-difference scheme for the phase field crystal equation. SIAM Journal on Numerical Analysis , 47(3):2269–2288, 2009

    2009

  2. [10]

    Numerical approximations of Allen-Cahn and Cahn–Hilliard equations

    Jie Shen and Xiaofeng Yang. Numerical approximations of Allen-Cahn and Cahn–Hilliard equations. Discrete Contin. Dyn. Syst , 28(4):1669–1691, 2010

    2010

  3. [11]

    On linear schemes for a Cahn–Hilliard diffuse interface model.Journal of Computational Physics , 234:140–171, 2013

    Francisco Guill´ en-Gonz´ alez and Giordano Tierra. On linear schemes for a Cahn–Hilliard diffuse interface model.Journal of Computational Physics , 234:140–171, 2013

    2013

  4. [12]

    Linear, first and second-order, unconditionally energy stable numerical schemes for the phase field model of homopolymer blends

    Xiaofeng Yang. Linear, first and second-order, unconditionally energy stable numerical schemes for the phase field model of homopolymer blends. Journal of Computational Physics , 327:294–316, 2016

    2016

  5. [13]

    A new class of efficient and robust energy stable schemes for gradient flows

    Jie Shen, Jie Xu, and Jiang Yang. A new class of efficient and robust energy stable schemes for gradient flows. SIAM Review, 61(3):474–506, 2019

    2019

  6. [14]

    Convergence and error analysis for the scalar auxiliary variable (SA V) schemes to gradient flows

    Jie Shen and Jie Xu. Convergence and error analysis for the scalar auxiliary variable (SA V) schemes to gradient flows. SIAM Journal on Numerical Analysis , 56(5):2895–2912, 2018

    2018

  7. [15]

    On large time-stepping methods for the Cahn–Hilliard equation

    Yinnian He, Yunxian Liu, and Tao Tang. On large time-stepping methods for the Cahn–Hilliard equation. Applied Numerical Mathematics, 57(5-7):616–628, 2007

    2007

  8. [16]

    On the stability and accuracy of partially and fully implicit schemes for phase field modeling

    Jinchao Xu, Yukun Li, Shuonan Wu, and Arthur Bousquet. On the stability and accuracy of partially and fully implicit schemes for phase field modeling. Computer Methods in Applied Mechanics and Engineering , 345:826–853, 2019

    2019

  9. [17]

    Effective time step analysis of a nonlinear convex splitting scheme for the cahn–hilliard equation

    Seunggyu Lee and Junseok Kim. Effective time step analysis of a nonlinear convex splitting scheme for the cahn–hilliard equation. Commun Comput Phys , 25(2):448–460, 2019

    2019

  10. [18]

    Convergence analysis of exponential time differencing schemes for the Cahn-Hilliard equation

    Xiao Li. Convergence analysis of exponential time differencing schemes for the Cahn-Hilliard equation. Communications in Computational Physics , 26(5), 2019

    2019

  11. [19]

    Energy-decreasing exponential time differencing Runge–Kutta methods for phase-field models

    Zhaohui Fu and Jiang Yang. Energy-decreasing exponential time differencing Runge–Kutta methods for phase-field models. Journal of Computational Physics , 454:110943, 2022

    2022

  12. [20]

    Unconditionally energy decreasing high-order Implicit-Explicit Runge-Kutta methods for phase-field models with the Lipschitz nonlinearity

    Zhaohui Fu, Tao Tang, and Jiang Yang. Unconditionally energy decreasing high-order Implicit-Explicit Runge-Kutta methods for phase-field models with the Lipschitz nonlinearity. arXiv preprint arXiv:2203.06034 , 2022

    2022 arXiv

  13. [21]

    The explicit–implicit–null method: Removing the numerical instability of PDEs

    Laurent Duchemin and Jens Eggers. The explicit–implicit–null method: Removing the numerical instability of PDEs. Journal of Computational Physics , 263:37–52, 2014

    2014

  14. [22]

    Stability and convergence analysis for the implicit-explicit method to the Cahn-Hilliard equation

    Dong Li, Chaoyu Quan, and Tao Tang. Stability and convergence analysis for the implicit-explicit method to the Cahn-Hilliard equation. Mathematics of Computation , 91(334):785–809, 2022

    2022

  15. [23]

    Implicit-explicit scheme for the Allen-Cahn equation preserves the maximum principle

    Tao Tang and Jiang Yang. Implicit-explicit scheme for the Allen-Cahn equation preserves the maximum principle. Journal of Computational Mathematics , pages 451–461, 2016

    2016

  16. [24]

    Maximum principle preserving exponential time differencing schemes for the nonlocal Allen–Cahn equation

    Qiang Du, Lili Ju, Xiao Li, and Zhonghua Qiao. Maximum principle preserving exponential time differencing schemes for the nonlocal Allen–Cahn equation. SIAM Journal on numerical analysis , 57(2):875–898, 2019

    2019

  17. [25]

    Douglas Lawson

    J. Douglas Lawson. Generalized Runge-Kutta Processes for Stable Systems with Large Lipschitz Constants. SIAM Journal on Numerical Analysis , 4(3):372–380, 1967

    1967

  18. [26]

    Strong stability preserving integrating factor Runge–Kutta methods

    Leah Isherwood, Zachary J Grant, and Sigal Gottlieb. Strong stability preserving integrating factor Runge–Kutta methods. SIAM Journal on Numerical Analysis , 56(6):3276–3307, 2018

    2018

  19. [27]

    Removing the stiffness from interfacial flows with surface tension

    Thomas Y Hou, John S Lowengrub, and Michael J Shelley. Removing the stiffness from interfacial flows with surface tension. Journal of Computational Physics , 114(2):312–338, 1994. Title Suppressed Due to Excessive Length 29

    1994

  20. [28]

    A diffuse interface model for microstructural evolution in elastically stressed solids

    Perry H Leo, John S Lowengrub, and Herng-Jeng Jou. A diffuse interface model for microstructural evolution in elastically stressed solids. Acta materialia, 46(6):2113–2130, 1998

    1998

  21. [29]

    Third-order conservative sign-preserving and steady-state-preserving time integrations and applications in stiff multispecies and multireaction detonations

    Jie Du and Yang Yang. Third-order conservative sign-preserving and steady-state-preserving time integrations and applications in stiff multispecies and multireaction detonations. Journal of Computational Physics , 395:489–510, 2019

    2019

  22. [30]

    Bound-preserving modified exponential Runge–Kutta discontinuous Galerkin methods for scalar hyperbolic equations with stiff source terms

    Juntao Huang and Chi-Wang Shu. Bound-preserving modified exponential Runge–Kutta discontinuous Galerkin methods for scalar hyperbolic equations with stiff source terms. Journal of Computational Physics , 361:111–135, 2018

    2018

  23. [31]

    Explicit third-order unconditionally structure- preserving schemes for conservative Allen–Cahn equations

    Hong Zhang, Jingye Yan, Xu Qian, Xiaowei Chen, and Songhe Song. Explicit third-order unconditionally structure- preserving schemes for conservative Allen–Cahn equations. Journal of Scientific Computing , 90:1–29, 2022

    2022

  24. [32]

    Spectral methods: algorithms, analysis and applications , volume 41

    Jie Shen, Tao Tang, and Li-Lian Wang. Spectral methods: algorithms, analysis and applications , volume 41. Springer Science & Business Media, 2011

    2011

  25. [33]

    Strong stability preserving Runge-Kutta and multistep time discretizations

    Sigal Gottlieb, David I Ketcheson, and Chi-Wang Shu. Strong stability preserving Runge-Kutta and multistep time discretizations. World Scientific, 2011

    2011

  26. [34]

    Third-order accurate, large time-stepping and maximum-principle-preserving schemes for the Allen-Cahn equation

    Hong Zhang, Xu Qian, and Songhe Song. Third-order accurate, large time-stepping and maximum-principle-preserving schemes for the Allen-Cahn equation. Numerical Algorithms, pages 1–38, 2023

    2023

  27. [35]

    Solving ordinary differential equations I: Nonstiff problems

    Ernst Hairer, Syvert P Nørsett, and Gerhard Wanner. Solving ordinary differential equations I: Nonstiff problems . Springer-Verlag, 1993

    1993

  28. [36]

    Characterizing the stabilization size for semi-implicit Fourier-spectral method to phase field equations

    Dong Li, Zhonghua Qiao, and Tao Tang. Characterizing the stabilization size for semi-implicit Fourier-spectral method to phase field equations. SIAM Journal on Numerical Analysis , 54(3):1653–1681, 2016

    2016

  29. [37]

    Functions of matrices: theory and computation

    Nicholas J Higham. Functions of matrices: theory and computation . SIAM, 2008

    2008

  30. [38]

    Average energy dissipation rates of explicit exponential Runge-Kutta methods for gradient flow problems

    Honglin Liao and Xuping Wang. Average energy dissipation rates of explicit exponential Runge-Kutta methods for gradient flow problems. Mathematics of Computation , 2024

    2024

  31. [39]

    Higher-order energy-decreasing exponential time differencing Runge-Kutta methods for gradient flows

    Zhaohui Fu, Jie Shen, and Jiang Yang. Higher-order energy-decreasing exponential time differencing Runge-Kutta methods for gradient flows. arXiv preprint arXiv:2402.15142 , 2024

    2024 arXiv

  32. [40]

    Exponential integrators

    Marlis Hochbruck and Alexander Ostermann. Exponential integrators. Acta Numerica, 19:209–286, 2010

    2010

  33. [41]

    A numerical method for the Cahn–Hilliard equation with a variable mobility.Communications in Nonlinear Science and Numerical Simulation , 12(8):1560–1571, 2007

    Junseok Kim. A numerical method for the Cahn–Hilliard equation with a variable mobility.Communications in Nonlinear Science and Numerical Simulation , 12(8):1560–1571, 2007

    2007

  34. [42]

    Coarsening kinetics from a variable-mobility Cahn-Hilliard equation: Application of a semi-implicit Fourier spectral method

    Jingzhi Zhu, Longqing Chen, Jie Shen, and Veena Tikare. Coarsening kinetics from a variable-mobility Cahn-Hilliard equation: Application of a semi-implicit Fourier spectral method. Physical Review E , 60(4):3564, 1999

    1999

  35. [43]

    A preconditioned steepest descent solver for the Cahn-Hilliard equation with variable mobility

    Xiaochun Chen, Cheng Wang, and S Wise. A preconditioned steepest descent solver for the Cahn-Hilliard equation with variable mobility. International journal of numerical analysis and modeling , 19(6), 2022

    2022

  36. [44]

    Stabilized crank-nicolson/adams-bashforth schemes for phase field models

    Xinlong Feng, Tao Tang, and Jiang Yang. Stabilized crank-nicolson/adams-bashforth schemes for phase field models. East Asian Journal on Applied Mathematics , 3(1):59–80, 2013

    2013

  37. [45]

    Efficient inequality-preserving integrators for differential equations satisfying forward Euler conditions

    Hong Zhang, Xu Qian, Jun Xia, and Songhe Song. Efficient inequality-preserving integrators for differential equations satisfying forward Euler conditions. ESAIM: Mathematical Modelling and Numerical Analysis , 57(3):1619–1655, 2023

    2023

  38. [46]

    Exponential time differencing for stiff systems

    Steven M Cox and Paul C Matthews. Exponential time differencing for stiff systems. Journal of Computational Physics , 176(2):430–455, 2002

    2002

  39. [47]

    Sobolev spaces

    Robert A Adams and John JF Fournier. Sobolev spaces. Elsevier, 2003

    2003

  40. [48]

    On the energy stability of Strang-splitting for Cahn-Hilliard

    Dong Li and Chaoyu Quan. On the energy stability of Strang-splitting for Cahn-Hilliard. arXiv preprint arXiv:2107.05349, 2021

    2021 arXiv

  41. [49]

    The operator-splitting method for Cahn-Hilliard is stable

    Dong Li and Chaoyu Quan. The operator-splitting method for Cahn-Hilliard is stable. Journal of Scientific Computing , 90:1–12, 2022

    2022

  42. [50]

    Explicit exponential runge–kutta methods for semilinear parabolic prob- lems

    Marlis Hochbruck and Alexander Ostermann. Explicit exponential runge–kutta methods for semilinear parabolic prob- lems. SIAM Journal on Numerical Analysis , 43(3):1069–1090, 2005

    2005

  43. [51]

    An adaptive time-stepping strategy for the Cahn-Hilliard equation

    Zhengru Zhang and Zhonghua Qiao. An adaptive time-stepping strategy for the Cahn-Hilliard equation. Communications in Computational Physics , 11(4):1261–1278, 2012

    2012

  44. [52]

    Molecular beam epitaxy: fundamentals and current status , volume 7

    Marian A Herman and Helmut Sitter. Molecular beam epitaxy: fundamentals and current status , volume 7. Springer Science & Business Media, 2012

    2012

  45. [53]

    Modeling elasticity in crystal growth

    KR Elder, Mark Katakowski, Mikko Haataja, and Martin Grant. Modeling elasticity in crystal growth. Physical review letters, 88(24):245701, 2002

    2002

Pith tools

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