arxiv: 2604.25520 · v2 · submitted 2026-04-28 · 🧮 math.FA

Recognition: unknown

Gamma-convergence, variational analysis and characterisation of minimisers for (s,p)-Gagliardo energies in the flat d-torus

F. Santilli, G. Pini

Pith reviewed 2026-05-07 14:30 UTC · model grok-4.3

classification 🧮 math.FA

keywords gamma-convergencegagliardo seminormfractional p-energyperiodic torusdirichlet energypiecewise affine functionsvariational analysisminimizers

0 comments

The pith

Rescaled (s,p)-Gagliardo energies on the torus Gamma-converge to a double-integral functional as s approaches 0 and to the classical p-Dirichlet energy as s approaches 1.

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

The paper establishes Gamma-convergence results for the density of the (s,p)-Gagliardo seminorm in spaces of L^p periodic functions on the flat d-torus. As s tends to 0 from above, the rescaled energy s times the seminorm density converges in the Gamma sense to a functional given by the double integral over the cell of |u(x) minus u(y) to the p|. As s tends to 1 from below, the rescaled energy (1 minus s) times the seminorm density Gamma-converges to the standard Dirichlet integral of the gradient to the p. In one dimension these limits are further analyzed inside a restricted class of piecewise affine periodic functions, where the energy reduces to a function of jump locations and is minimized when the jumps are equally spaced. These results matter because they connect fractional nonlocal energies to their local limits in a periodic setting where boundary effects are absent.

Core claim

We prove that as s to 0+, s F_p^s Gamma-converges to F_p^0 given by the double integral of |u(x)-u(y)|^p over the periodic cell, while as s to 1-, (1-s) F_p^s Gamma-converges to the classical Dirichlet p-energy. Within the special class of one-dimensional piecewise affine periodic functions whose distributional derivative has a constant absolutely continuous part and a singular part consisting of opposite-sign quantized jumps, the energy of both F_p^s and F_p^0 depends only on the positions of the jumps and attains its absolute minimum at the equispaced configuration.

What carries the argument

The rescaled (s,p)-Gagliardo seminorm density F_p^s on L^p periodic functions, together with the restricted class of piecewise affine periodic functions whose distributional derivative splits into a constant absolutely continuous part and a singular part of opposite sign with quantized jumps.

If this is right

The Gamma-limit functionals admit explicit integral representations that can be used directly for minimization problems.
Existence of minimizers follows immediately from the Gamma-convergence and lower semicontinuity properties in the periodic L^p setting.
In one dimension the energy depends only on jump locations, reducing the variational problem to a finite-dimensional optimization over point configurations.
The equispaced configuration is independent of both p and s inside the restricted function class, giving a uniform minimizer across the family of energies.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same rescaling arguments may extend the Gamma-convergence statements to other compact manifolds without boundary.
The reduction of the energy to jump positions in the piecewise affine class suggests that lattice or dislocation models could be recovered as s tends to zero.
Numerical schemes that discretize the double-integral functional could serve as approximations to the fractional energies for small s.
The quantized-jump restriction may be relaxed by density arguments if the Gamma-convergence is shown to be compatible with approximation by such functions.

Load-bearing premise

The Gamma-convergence statements hold for functions in L^p periodic spaces, while the minimizer characterization requires restricting attention to the subclass of piecewise affine periodic functions whose distributional derivative has a constant absolutely continuous part and a singular part with opposite sign and quantized jumps.

What would settle it

An explicit computation or numerical evaluation showing that a configuration of jumps that are not equally spaced produces strictly lower energy than the equispaced configuration for the limit functional F_p^0 would falsify the one-dimensional minimizer claim.

Figures

Figures reproduced from arXiv: 2604.25520 by F. Santilli, G. Pini.

**Figure 1.** Figure 1: The key observation is that a vector I ∈ Z d always points to the corner with smallest components of the hypercube QI . This means that a hypercube in a sector with any negative coordinate is “closer” to the origin than a hypercube in the sector with all positive coordinates having a vector of the same norm. We need to consider this fact to not over/under-estimate [u]s,p|Bc R . In the figure above the vect… view at source ↗

**Figure 2.** Figure 2: A visualisation of edge dislocations in a square crystal lattice with two different atomic distances. The image represents the view of a section orthogonal to the x3 axis. Following the general idea of [13, Section 2]), let H± := R × R ± × R and H0 := R × {0} × R. According to isotropic linear elasticity, the elastic energy of a planar displacement U ± ∈ H1 (H±, R 3 ) (where U := U ± on H±) is given by E e… view at source ↗

**Figure 3.** Figure 3: In the scheme above we picture dislocation as “lines” orthogonal to the (x1, x2)-plane in the direction x ′ , representing the last d − 2 variables. In this setting the multivector t tangent to the dislocation manifold M is pointing in the direction of x ′ . In this framework, the deformation of the crystal is modelled by the absolutely continuous part of the distributional displacement gradient, represent… view at source ↗

**Figure 4.** Figure 4: This figure is an example of a rigid variation applied to a function u[X] with a jump in xi . The effect of the variation on the function is the same as moving the jump from xi to xi + h. where u[X] is the T-periodic function defined (up to vertical translations) by the relation Du[X] = L 1 − X xi∈X δxi =: L 1 − µ X. We observe that the energy F s p [X] = F s p [X+a], for every a ∈ R, where X+a = (xn+a)n. … view at source ↗

**Figure 5.** Figure 5: The objects in the proof of Proposition 3.8. Now, taking the limit as h → 0, we compute lim h→0 + F s p [X + hei ] − Fs p [X] h = lim σ→0 + 2 ˆ R\Bσ (xi) |u(xi) − u(y) + 1| p − |u(xi) − u(y)| p |xi − y| 1+sp dy, where on 1 hAh 1 we used the Mean Value Theorem as h → 0 +, arriving to equation (3.8). Now, we want to prove equation (3.9). To do so, we consider a rigid variation in the direction of ej where i … view at source ↗

**Figure 6.** Figure 6: This figure is an example of a rigid variation applied to a regularised function u ε [X] with a jump in xi . The effect of the variation on the function is the same as moving the regularised transition from xi to xi + h. To relieve the notation we will denote by ρ i ε (x) := ρε(x − xi) and with ρ˜ i ε (x) := P k∈Z ρε(x − xi − kT). Proposition 3.15 (First and second variations for F s,ε p ). Let p > 1, 1 p … view at source ↗

**Figure 7.** Figure 7: An example of how the measure µ X behaves for a particular choice of X and a fixed t. For convenience the set At(1) has been highlighted. Now, by taking At(k) := {x ∈ [0, T)| µ X([x, x + t]) = k}, we apply the change of variables y = x + ε and t = h − 2ε in equation (3.29), using periodicity of µ X, yielding F s,ε p [X] ≥ 2 2−p ˆ R−2ε 0 1 (t + 2ε) 2 ˆ T 0 |µ X([y, y + t])| pdydt − 2Rp−1T p − 1 = 22−p ˆ R−2… view at source ↗

read the original abstract

This paper deals with the variational analysis, for every $s \in (0,1)$ and $p \in [1,+\infty)$, of $(s,p)$-Gagliardo seminorms in a periodic setting. First, we consider the space of $L^p$, $T$-periodic functions and define the energy functional $\mathcal{F}_p^s$ as the density of the $d$-dimensional $(s,p)$-Gagliardo seminorm over the periodic cell. Our goal is to rigorously characterise the $\Gamma$-limits of this functional as the fractional parameter $s$ approaches its endpoint values, $0^+$ and $1^-$. We prove that, as $s \to 0^+$, the rescaled energy $s\mathcal{F}_p^s$ $\Gamma$-converges to a functional $\mathcal{F}_p^0$ defined by the double integral of $|u(x)-u(y)|^p$ over the periodic cell. Then, for the limit as $s \to 1^-$, we establish that the rescaled energy $(1-s)\mathcal{F}_p^s$ $\Gamma$-converges to the classical Dirichlet $p$-energy, extending known results from bounded domains to the periodic framework. Finally, we analyse the one-dimensional minimiser of the energy $\mathcal{F}_p^s$ for $s \in (0,1)$ and the limit functional $\mathcal{F}_p^0$ within the special class of piecewise affine periodic functions whose distributional derivative consists of a constant absolutely continuous part and a singular part with opposite sign and quantised jumps. In this setting, the energy depends only on the position of these jump points, and we prove that the absolute minimiser is achieved by their equispaced configuration.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Extends Gamma-convergence for the rescaled Gagliardo energies to the periodic torus and pins down equispaced minimizers inside a narrow 1D piecewise-affine class, but the absolute claim rests on an unverified restriction.

read the letter

The main contributions are the two Gamma-convergence statements on the torus and the explicit 1D minimizer result inside a restricted function class. The paper shows that s F_p^s Gamma-converges to the double-integral functional F_p^0 as s approaches 0 from above, and that (1-s) F_p^s Gamma-converges to the classical Dirichlet p-energy as s approaches 1 from below. Both limits are stated for L^p periodic functions on the flat d-torus. It also reduces the 1D minimization of F_p^s and F_p^0 to the locations of quantized jumps for piecewise-affine functions whose distributional derivative has a constant absolutely continuous part plus a singular part of opposite sign, and proves the equispaced configuration wins inside that class. These are legitimate extensions of existing Gamma-convergence work from bounded domains, and the 1D calculation is concrete because the energy depends only on jump positions once the class is fixed. The definitions and statements are clear, and the periodic setting is handled without obvious technical gaps in the abstract claims. The soft spot is the minimizer statement. The abstract calls the equispaced configuration the absolute minimiser, yet the result is proved only inside the special piecewise-affine subclass. No density, relaxation, or lower-semicontinuity argument is indicated that would show the global infimum over all L^p functions is attained there or that competitors outside the class cannot do better. If the full paper does not supply that bridge, the wording overstates the scope. The Gamma-convergence parts themselves appear to be standard adaptations and do not carry the same restriction. This paper is for specialists in fractional Sobolev spaces and variational analysis who already work with Gagliardo seminorms or periodic problems. A reader following that literature will get a clean extension plus one explicit example. It deserves a serious referee because the core limits are well-posed extensions and the 1D result, even if scoped to the subclass, is a tangible addition. I would send it to peer review and expect the referees to ask for clarification on whether the minimizer is global or only relative to the chosen class.

Referee Report

1 major / 2 minor

Summary. The paper studies (s,p)-Gagliardo seminorms on the flat d-torus for s in (0,1) and p in [1,infty). It defines the periodic energy functional F_p^s as the density of the d-dimensional (s,p)-Gagliardo seminorm over the periodic cell in L^p(T^d). It proves that s F_p^s Gamma-converges to the functional F_p^0 given by the double integral of |u(x)-u(y)|^p over the cell as s->0+, and that (1-s) F_p^s Gamma-converges to the classical Dirichlet p-energy as s->1-. In one dimension, within the restricted class of piecewise-affine 1-periodic functions whose distributional derivative consists of a constant absolutely continuous part plus a singular part with opposite sign and quantized jumps, it shows that the energy depends only on jump locations and that the equispaced configuration is the absolute minimizer for both F_p^s and F_p^0.

Significance. If the Gamma-convergence statements hold in full L^p(T^d), the results extend standard fractional-to-local and fractional-to-nonlocal limits to the periodic setting, which is relevant for homogenization and periodic variational problems. The explicit characterization of minimizers inside the indicated piecewise-affine subclass provides concrete insight into the dependence on jump positions, though the global validity of the minimizer claim requires additional justification.

major comments (1)

[analysis of one-dimensional minimisers (final section)] The claim that the equispaced configuration achieves the absolute minimizer of F_p^s (for s in (0,1)) and of F_p^0 is established only inside the restricted subclass of piecewise-affine 1-periodic functions whose distributional derivative equals a constant AC part plus a singular measure with opposite sign and quantized jumps. No density, relaxation, or lower-semicontinuity argument is supplied showing that the infimum of F_p^s or F_p^0 over all of L^p(T) is attained inside this subclass, nor that competitors outside it cannot achieve strictly lower energy. This gap directly affects the 'absolute' qualifier in the final claim.

minor comments (2)

[Introduction and definition of F_p^s] The precise definition of the periodic cell and the normalization factor in the density defining F_p^s should be stated explicitly with an equation number already in the introduction or §2, to avoid ambiguity when comparing with the non-periodic literature.
[Gamma-convergence theorems] In the Gamma-convergence statements, clarify whether the convergence is with respect to the L^p topology or a weaker topology, and whether the limit functionals are defined on the same space or require extension by lower semicontinuity.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading and constructive feedback on our manuscript. We address the major comment below.

read point-by-point responses

Referee: [analysis of one-dimensional minimisers (final section)] The claim that the equispaced configuration achieves the absolute minimizer of F_p^s (for s in (0,1)) and of F_p^0 is established only inside the restricted subclass of piecewise-affine 1-periodic functions whose distributional derivative equals a constant AC part plus a singular measure with opposite sign and quantized jumps. No density, relaxation, or lower-semicontinuity argument is supplied showing that the infimum of F_p^s or F_p^0 over all of L^p(T) is attained inside this subclass, nor that competitors outside it cannot achieve strictly lower energy. This gap directly affects the 'absolute' qualifier in the final claim.

Authors: We appreciate the referee highlighting this point. The analysis in the final section is carried out exclusively within the indicated subclass of piecewise-affine 1-periodic functions whose distributional derivative consists of a constant absolutely continuous part plus a singular part with opposite sign and quantized jumps. Within this class we prove that the energy depends only on the jump locations and that the equispaced configuration is the minimizer. The manuscript does not claim, nor provide arguments to establish, that the result extends to a global minimizer over all of L^p(T); no density, relaxation or lower-semicontinuity arguments are supplied for this purpose. The term 'absolute' is used to indicate the minimizer inside the considered subclass. To eliminate any possible ambiguity, we will revise the abstract and the statements in the final section to explicitly qualify the minimality result as holding within this subclass. This is a clarification that preserves the scope and validity of the presented results. revision: yes

Circularity Check

0 steps flagged

No circularity; standard Gamma-convergence proofs extended directly to periodic setting

full rationale

The paper explicitly defines F_p^s as the density of the (s,p)-Gagliardo seminorm on the periodic cell and derives the two Gamma-limits (s F_p^s to the double-integral functional as s→0+ and (1-s) F_p^s to the Dirichlet energy as s→1-) via standard variational techniques in the full L^p(T^d) space. The minimizer characterization is stated only inside the explicitly restricted subclass of piecewise-affine functions with the given derivative structure; no global extension or density argument is invoked that would collapse back to the definition. No self-definitional loops, fitted inputs renamed as predictions, or load-bearing self-citations appear in the derivation chain.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The work relies entirely on standard definitions and techniques from functional analysis and Gamma-convergence theory without introducing new free parameters, ad-hoc axioms, or invented entities.

axioms (1)

standard math Standard definitions of (s,p)-Gagliardo seminorms, Gamma-convergence, and L^p periodic function spaces
Invoked throughout the abstract to state the energies and limits.

pith-pipeline@v0.9.0 · 5645 in / 1355 out tokens · 65614 ms · 2026-05-07T14:30:31.011345+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

32 extracted references · 2 canonical work pages

[1]

Ambrosio

V. Ambrosio. On some convergence results for fractional periodic Sobolev spaces.Opuscula Math., 40(1):5–20, 2020

2020
[2]

Bourgain, H

J. Bourgain, H. Brezis, and P. Mironescu. Another look at Sobolev spaces.Optimal Control and Partial Differential Equations, pages 439–455, 2001

2001
[3]

Bourgain, H

J. Bourgain, H. Brezis, and P. Mironescu. Limiting embedding theorems forWs,p whens→1and applications. J. Anal. Math., 87(1):77–101, 2002

2002
[4]

Braides.Gamma-Convergence for Beginners

A. Braides.Gamma-Convergence for Beginners. Oxford Lecture Series in Mathematics and Its Applications. Clarendon Press, Oxford, England, 2002

2002
[5]

Brasco, E

L. Brasco, E. Parini, and M. Squassina. Stability of variational eigenvalues for the fractionalp−Laplacian. Discrete and Continuous Dynamical Systems, 36(4):1813–1845, 2015

2015
[6]

HBrezis.Howtorecognizeconstantfunctions.connectionswithSobolevspaces.Russian Mathematical Surveys, 57(4):693–708, 2007

2007
[7]

Brezis.Functional analysis, Sobolev spaces and partial differential equations

H. Brezis.Functional analysis, Sobolev spaces and partial differential equations. Universitext. Springer, New York, NY, 2010

2010
[8]

Cacace and A

S. Cacace and A. Garroni. A multi-phase transition model for dislocations with interfacial microstructure. Interfaces Free Bound., 11(2):291–316, 2009

2009
[9]

Caffarelli and L

L. Caffarelli and L. Silvestre. An extension problem related to the fractional Laplacian.Comm. Partial Dif- ferential Equations, 32(8):1245–1260, 2007

2007
[10]

Crismale, L

V. Crismale, L. De Luca, A. Kubin, A. Ninno, and M. Ponsiglione. The variational approach tos-fractional heat flows and the limit casess→0+ ands→1 −.J. Funct. Anal., 284(8), 2023

2023
[11]

De Luca, A

L. De Luca, A. Garroni, and M. Ponsiglione.Γ-convergence analysis of systems of edge dislocations: The self energy regime.Arch. Ration. Mech. Anal., 206(3):885–910, 2012

2012
[12]

De Luca, M

L. De Luca, M. Goldman, and M. Ponsiglione. Dynamics of screened particles towards equi-spaced ground states, 2025

2025
[13]

De Luca, M

L. De Luca, M. Ponsiglione, and E. Spadaro. Two slope functions minimizing fractional seminorms and appli- cations to misfit dislocations.SIAM Journal on Mathematical Analysis, 56(1):1179–1196, 2024

2024
[14]

Di Nezza, G

E. Di Nezza, G. Palatucci, and Valdinoci E. Hitchhiker’s guide to the fractional Sobolev spaces.Bull. Sci. Math., 136:521–573, 2012

2012
[15]

Fanzon, M

S. Fanzon, M. Ponsiglione, and R. Scala. Uniform distribution of dislocations in Peierls–Nabarro models for semi-coherent interfaces.Calculus of Variations and Partial Differential Equations, 59(4), 2020

2020
[16]

Garroni and S

A. Garroni and S. Müller.Γ-limit of a phase-field model of dislocations.SIAM J. Math. Anal., 36(6):1943–1964, 2005

1943
[17]

Giuliani, J

A. Giuliani, J. L. Lebowitz, and E. H. Lieb. Striped phases in two-dimensional dipole systems.Physical Review B, 76(18), 2007

2007
[18]

Giuliani, J

A. Giuliani, J. L. Lebowitz, and E. H. Lieb. Periodic minimizers in 1d local mean field theory.Communications in Mathematical Physics, 286(1):163–177, 2008

2008
[19]

Giuliani and S

A. Giuliani and S. Müller. Striped periodic minimizers of a two-dimensional model for martensitic phase transitions.Communications in Mathematical Physics, 309(2):313–339, 2011

2011
[20]

Kubin, V

A. Kubin, V. Pagliari, and A. Tribuzio. Second-order asymptotics of fractional Gagliardo seminorms ass→1− and convergence of the associated gradient flows. 2024. arXiv:2410.17829

work page arXiv 2024
[21]

Kubin, G

A. Kubin, G. Saracco, and G. Stefani. On theΓ-limit of weighted fractional energies. 2025. arXiv:2503.19875

work page arXiv 2025
[22]

G.Leoni.A first course in fractional Sobolev spaces.GraduateStudiesinMathematics.AmericanMathematical Society, Providence, RI, 2023. 36 G. PINI, F. SANTILLI

2023
[23]

J. M. Mazón, J. D. Rossi, and J.Toledo. Fractional p-Laplacian evolution equations.Journal de Mathématiques Pures et Appliquées, 105(6):810–844, 2016

2016
[24]

Maz’ya and T

V. Maz’ya and T. Shaposhnikova. On the Bourgain, Brezis, and Mironescu Theorem concerning limiting embeddings of fractional Sobolev spaces.Journal of Functional Analysis, 195(2):230–238, 2002

2002
[25]

S. Müller. Singular perturbations as a selection criterion for periodic minimizing sequences.Calculus of Vari- ations and Partial Differential Equations, 1(2):169–204, 1993

1993
[26]

A.C. Ponce. A new approach to sobolev spaces and connections toΓ-convergence.Calc. Var. Partial Differ. Equ., 19(3):229–255, 2004

2004
[27]

W. T. Read and W. Shockley. Dislocation models of crystal grain boundaries.Phys. Rev., 78:275–289, 1950

1950
[28]

Ren and J

X. Ren and J. Wei. On energy minimizers of the diblock copolymer problem.Interfaces and Free Boundaries, Mathematical Analysis, Computation and Applications, 5(2):193–238, 2003

2003
[29]

Sandier and S

E. Sandier and S. Serfaty. From the Ginzburg-Landau model to vortex lattice problems.Communications in Mathematical Physics, 313(3):635–643, 2012

2012
[30]

Triebel.Interpolation theory, function spaces, differential operators

H. Triebel.Interpolation theory, function spaces, differential operators. North-Holland mathematical library. North-Holland Publishing, Oxford, New Zealand, 1978

1978
[31]

van der Merwe

J. van der Merwe. On the stresses and energies associated with inter-crystalline boundaries.Proceedings of the Physical Society. Section A, 63(6):616–637, 1950

1950
[32]

La Sapienza

Gao Y., J.-G. Liu, T. Luo, and Y. Xiang. Revisit of the Peierls-Nabarro model for edge dislocations in hilbert space.Discrete and Continuous Dynamical Systems - Series B, 22(11), 2020. (G. Pini)Dip. di Matematica, Univ. Roma-I “La Sapienza”, Piazzale Aldo Moro 5, 00185, Roma, Italy E-mail address, G. Pini:giovanni.pini@uniroma1.it (F. Santilli)Dip. di Mat...

2020