Existence and Uniqueness for the Fractional Gelfand Equation in $\mathbb{R}$

Enno Lenzmann; Florian P. Lanz

Existence and Uniqueness for the Fractional Gelfand Equation in $\mathbb{R}$

T0 review · 0 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Fractional Gelfand equation has exactly one solution shape

desk verdict Completes the classification for s in (1/2,1) with a novel fixed-point proof; solid, minor external-input caveats, worth a serious referee. read the letter →

arxiv 2506.07577 v1 pith:RVPTVLYE submitted 2025-06-09 math.AP

Florian P. Lanz , Enno Lenzmann This is my paper

classification math.AP MSC 35R1135A0135A0235B06

keywords fractionalGelfandequationLaplaciannonlocalLiouvilleuniquenessfixedpointmethodreverseHardy-Littlewood-SobolevinequalityconjugateRieszpotentialfiniteMorseindex

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 that the one-dimensional fractional Gelfand equation $(-\Delta)^s u = e^u$ with finite mass $\int_{\mathbb{R}} e^u\,dx < \infty$ has, for every $s\in(\tfrac12,1)$, exactly one solution profile up to translation and scaling; the profile is smooth, even, and strictly decreasing. Existence is obtained by rewriting the equation with $v=\sqrt{e^u}$ and finding fixed points of a nonlocal shooting map. The proof passes through a Gaussian-regularized version of the problem and takes a careful limit $\sigma\to 0$ using the reverse Hardy-Littlewood-Sobolev inequality at the conformally invariant endpoint. Uniqueness follows from a nondegeneracy result for the linearized map, and the same machinery yields analogous results for a class of positive, even, monotone-decreasing weights $K(x)$.

What carries the argument

The argument is carried by a fixed point scheme in the variable $v=\sqrt{e^u}$, with the conjugate Riesz potential $H_\alpha=H\circ(-\Delta)^{-\alpha}$, $\alpha=s-\tfrac12$, converting the equation into the first-order nonlocal ODE $\partial_x v=-\tfrac12 H_\alpha(v^2)v$. Fixed points of $T_\lambda[v](x)=\lambda\sqrt{K(x)}\exp(-\tfrac12\int_0^x H_\alpha(v^2)\,dy)$ are found first with a Gaussian weight $e^{-\sigma^2 x^2}$ using Schaefer-Schauder fixed point theory, and the limit $\sigma\to 0$ is controlled by Pohozaev-type identities plus the reverse Hardy-Littlewood-Sobolev inequality in the conformally invariant case $q=1/(1+\alpha)$, which supplies $\sigma$-independent $L^p$ bounds. Uniqueness rests on a nondegeneracy lemma: the linearized fixed point operator has no eigenvalue $1$, proved by a positivity argument that represents a certain integral as the $L^2$ norm of a one-sided Laplace transform.

What would settle it

Compute or look up the sharp constant in the reverse Hardy-Littlewood-Sobolev inequality (2.6) at $\beta=0$, $q=1/(1+\alpha)$ for $\alpha\in(0,\tfrac12)$; a counterexample pair $\rho\in L^1\cap L^q$ violating the inequality would break Lemma 5.2 and the $\sigma\to 0$ existence step. Alternatively, numerically shoot fixed points of $T_\lambda$ for $s=\tfrac34$, $\lambda=1$; two distinct even profiles with $v(0)=1$ would refute Theorem 1.3.

Watch

Extended reading notes

Core claim

The central discovery is Theorem 1.3: for every $s\in(\tfrac12,1)$, there exists a unique $Q_s\in L_s(\mathbb{R})$ such that every solution $u\in L_s(\mathbb{R})$ of $(-\Delta)^s u=e^u$ with $e^u\in L^1(\mathbb{R})$ is $Q_s$ up to the symmetry $u(x)\mapsto u(\mu(x+y))+2s\log\mu$. The profile $Q_s$ is smooth, even, and strictly monotone-decreasing in $|x|$. Along with this, the paper establishes that every such solution has finite Morse index and is stable outside a compact set, and that the linearized operator $(-\Delta)^s-e^u$ has kernel spanned exactly by the two symmetry modes $\partial_x u$ and $x\partial_x u+2s$.

Load-bearing premise

The proof relies on the reverse Hardy-Littlewood-Sobolev inequality in its conformally invariant endpoint $q=1/(1+\alpha)$, cited rather than proved, to obtain the uniform $L^p$ control that lets Gaussian-regularized fixed points pass to the original equation; if that endpoint inequality fails for any $\alpha\in(0,\tfrac12)$, the existence argument collapses.

Editorial extensions

If this is right

Any solution of $(-\Delta)^s u=e^u$ with finite mass is smooth, even about some point, and strictly decreasing away from it.
The linearized operator is nondegenerate: all kernel elements are the translation and scaling modes.
All solutions have finite Morse index, so each is stable outside a compact set; no fully stable solution exists.
The uniqueness extends to $\int_{\mathbb{R}} K e^u\,dx<\infty$ for positive, even, monotone-decreasing $K$, with even solutions determined by their value at $0$.
The same uniqueness proof can be adapted to $s=\tfrac12$, giving another proof of the classical nonlocal Liouville uniqueness.

Reading between the lines

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

The Laplace-transform proof of nondegeneracy suggests a spectral reading: the kernel of the linearized operator corresponds to a Perron-Frobenius eigenvalue problem on the half-line, which may extend to higher-dimensional radial problems with explicit kernels.
The $\sigma\to 0$ selection mechanism indicates that Gaussian regularization plus the conformal endpoint of the reverse Hardy-Littlewood-Sobolev inequality forces uniform mass and decay; the same template may apply to other nonlocal equations with exponential nonlinearities.
A concrete numerical consequence is that shooting fixed points of $T_\lambda$ for fixed $s$ and varying $\lambda=v(0)$ should produce a single branch; an observed bifurcation would signal a failure of the nondegeneracy lemma.
Because the authors note that fast-decaying $K$ may admit linear-growth solutions, the monotone non-decaying class of weights is likely sharp for the stated uniqueness result.

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 a high-quality, careful paper that solves a real open problem in the one-dimensional fractional Gelfand equation. It deserves a serious referee and likely acceptance after minor revision.

The genuinely new result is Theorem 1.3: for every s in (1/2,1), there is exactly one finite-mass solution profile modulo translation and scaling. The s=1/2 case was known via conformal invariance; the non-conformal range needed new tools. The fixed-point reformulation with v = sqrt(e^u), the conjugate Riesz potential, the Gaussian regularization with sigma->0 limit, and the continuation uniqueness argument are all substantial. The proofs are detailed and internally consistent. The paper also establishes symmetry, smoothness, finite Morse index, and nondegeneracy, which round out the picture. The extension to nonconstant K under Assumption (A) is a useful bonus.

The soft spots are minor. The reverse Hardy-Littlewood-Sobolev inequality is used at the conformal endpoint q=1/(1+alpha) in Lemma 5.2 without proof, and the sigma-independent L^p bounds that carry the existence argument hang on it. That is the one load-bearing external input, and it is cited to four references. The endpoint is correctly stated - q = 2/(2+2alpha) matches the needed q - so I don't think the concern is fatal. A referee should ask for a precise statement with the exact endpoint conditions, or a proof in an appendix, but this is not a reason to doubt the result. Similarly, the negative-eigenvalue bound from [3] is cited without proof; that's standard. There is a small normalization slip in the proof of Theorem 1.3 where the text sets u(0)=0 but then defines Q_s with Q_s(0)=1; this is a cosmetic error in exposition, not a mathematical gap.

The citation pattern is healthy. Self-citations are to technique in [1] and to the upcoming thesis [17], neither of which states the main theorem. The circularity burden is low.

Who is this for? Specialists in nonlinear PDEs, fractional Laplacian, and Liouville-type classification problems. Anyone working on fractional Gelfand equations will need to cite it.

My verdict: accept for peer review. The central argument holds up, the soft spots are minor, and the result is new and complete.

Referee Report

0 major / 5 minor

Summary. This paper studies the one-dimensional fractional Gelfand equation (-Δ)^s u = e^u with e^u ∈ L^1(R) for s ∈ (1/2,1). The main results are that all such solutions are smooth, even after a translation, strictly monotone in |x|, and unique up to the scaling/translation symmetry u(x) ↦ u(µ(x+y)) + 2s log µ; in addition, every solution has finite Morse index and the linearized operator is nondegenerate. The proof rewrites the equation in terms of v = √(e^u), leading to the nonlocal first-order equation ∂_x v = -1/2 H_α(v²)v and a fixed point scheme T_λ. A Gaussian regularization T_λ^(σ) provides compactness, and a careful σ→0 limit using a reverse Hardy–Littlewood–Sobolev inequality gives existence for the original equation. Uniqueness is obtained through a nondegeneracy lemma and a global continuation argument, and the final sections extend the results to a class of nonconstant positive, even, monotone kernels K(x).

Significance. If correct, the paper gives a complete classification of finite-mass solutions to the fractional Gelfand equation in the supercritical range s > 1/2, which is genuinely new: the endpoint s = 1/2 had been understood via conformal invariance, but the present range requires different tools. The nonlocal shooting method, the use of the conjugate Riesz potential H_α, and the reverse HLS inequality at the conformal endpoint are interesting and likely to be influential. The paper is carefully written and the central arguments are internally consistent; the proofs of existence, including the delicate passage σ_n → 0, and of uniqueness via continuation and Perron–Frobenius arguments are detailed. The main external inputs are the reverse HLS inequality and a spectral bound from [3]; both are cited rather than proved, which is acceptable if the precise statements are included, but should be made fully transparent.

minor comments (5)

[7.2] There is a notation inconsistency: Section 1.2 and Theorem 1.5 define T_u = ∂_x u and R_u = x∂_x u + 2s, but the proof of Theorem 1.5 uses R = ∂_x u and T = x∂_x u + 2s. Since the final span is the same, this is only a presentation issue, but the notation should be aligned.
[5.2, Eq. (5.2)] The text says 'multiplying both sides with -xv', but the displayed identity corresponds to multiplication by -2xv. Please correct the factor or the wording.
[2.1 / Lemma 5.2] The reverse HLS inequality (2.6)/(5.6) is cited rather than proved. In the present application, with µ = 2α and q = 1/(1+α), the condition q > 1/(1+µ) holds and β = 0 is the conformally invariant endpoint, so the concern that the endpoint might be misstated does not land on reading the paper; nevertheless, please add a precise statement of the cited theorem, for example from [4] or [21], so the reader can verify that the endpoint case is exactly covered.
[7.1, Proposition 1.1] The finite-Morse-index bound imports [3, Theorem 1.1], which appears to be a recent preprint. Please state the theorem and explicitly verify its hypotheses for V = e^u, in particular that V ∈ L^1(R; |x|^{2α} dx) ∩ L^∞(R), so the reliance is transparent.
[5.1, Proposition 5.1] For σ ≠ 0, the proof invokes Lemma 4.1, but the equation solved by u = log(K^{-1}v²) is (-Δ)^s u = e^{-σ²x²}K e^u, not (1.1). The modified kernel e^{-σ²x²}K also satisfies Assumption (A), so the conclusion is unaffected, but this should be said explicitly.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the main theorems are derived from the equation via a self-contained fixed-point and continuation argument, with only external inequalities and non-load-bearing self-citations.

full rationale

The derivation chain is self-contained in the sense relevant to circularity. Existence (Theorem 1.2) is obtained from fixed points of the map T_lambda, first for the Gaussian-regularized map T_lambda^(sigma) and then by passing sigma to 0. The sigma-independent L^p bounds in Lemma 5.2 use the reverse Hardy-Littlewood-Sobolev inequality (2.6) and its conformal endpoint form (5.6); this inequality is imported from independent references [4,2,21,7] and is an external input, not an assumption equivalent to the theorem. Uniqueness (Theorem 1.3) rests on the nondegeneracy Lemma 6.2, which is proved directly by a Laplace-transform and Fredholm argument, and on the continuation argument in Lemma 6.3; it does not invoke Theorem 1.3 itself. The only self-references are [1] as the source of the v = sqrt(e^u) change of variables and moving-plane technique, [17] for an s = 1/2 adaptation, and [18] for a companion parabolic problem; none of these states or forces the main theorem. Theorem 1.5 uses Theorem 1.3 to reduce to Q_s, but this is a forward use of an already-proved theorem, and Theorem 1.3 did not depend on Theorem 1.5. The parameter lambda is a shooting/initial-value parameter fixed by the equation, not a fitted constant, and uniqueness up to translation and scaling is proved rather than assumed. Even if the cited reverse HLS endpoint inequality were later found to require extra hypotheses, that would be a correctness risk about an external result, not a circularity in the paper's own derivation chain.

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

No fitted constants and no new entities appear. The proof depends on standard external results: the reverse Hardy-Littlewood-Sobolev inequality, the bound on negative eigenvalues of fractional Schrödinger operators, the affinity of s-harmonic functions, Schauder regularity, and a non-existence theorem for stable solutions. Assumption (A) on K is an explicit domain assumption for the generalization; for K ≡ 1 it is satisfied trivially. The parameter λ is the value of the auxiliary function at zero, which is later removed by the scaling symmetry.

assumptions (6)

standard math Reverse Hardy-Littlewood-Sobolev inequality, conformally invariant endpoint q = 1/(1+α), stated in (2.6).
Used in Lemma 5.2 to obtain σ-independent L^p bounds for fixed points; cited to [4,2,21,7] and not proved in the paper.
standard math Bound on the number of negative eigenvalues of fractional Schrödinger operators: N_{<0}(H) ≤ C_s(∥|x|^{2α}V∥_{L¹} + 1) from [3, Theorem 1.1].
Used in Proposition 1.1 and Theorem 1.6(v) to prove finite Morse index via comparison with a Schrödinger operator.
standard math Entire s-harmonic functions in L_s(R) are affine (Fall [9]).
Used in Proposition 3.1 to characterize the kernel of (-Δ)^s and to identify the integral solution representation.
standard math Schauder-type regularity estimates for the fractional Poisson equation (adaptation of [20, Theorem 10.3]).
Used in Lemmas 3.2, 3.4 and the proof of Theorem 1.1 to bootstrap regularity of solutions.
standard math Nonexistence of stable solutions of (-Δ)^s u = e^u in R^N for N < 10s (Duong-Nguyen [8]).
Used in Theorem 1.4 to conclude that the Morse index of any solution is at least one.
domain assumption Assumption (A) on K: K > 0, even, monotone-decreasing in |x|, C¹ with ∂_x sqrt(K) in L^∞.
Assumed for the fixed-point framework and for Theorem 1.6; satisfied trivially by the constant case K ≡ 1.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Existence and Uniqueness for the Fractional Gelfand Equation in $\mathbb{R}$." pith.science (2026). https://pith.science/paper/RVPTVLYE

@misc{pith2026250607577,
  author       = {Pith},
  title        = {Pith review of: Existence and Uniqueness for the Fractional Gelfand Equation in $\mathbbR$},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RVPTVLYE}},
  note         = {Machine review of arXiv:2506.07577}
}

abstract

We prove existence, symmetry and uniqueness of solutions to the fractional Gelfand equation $$ (-\Delta)^s u = e^u \quad \mbox{in $\mathbb{R}$} \quad \mbox{with} \quad \int_{\mathbb{R}} e^u dx < +\infty $$ for all exponents $s \in (\frac{1}{2},1)$. Furthermore, we show $u$ has finite Morse index and that its linearized operator is nondegenerate. Our arguments are based on a fixed point scheme in terms of the function $v= \sqrt{e^u}$ and we devise a nonlocal shooting method involving (locally) compact nonlinear maps. We also study existence, symmetry and uniqueness of solutions to $(-\Delta)^s u = K e^u$ in $\mathbb{R}$ with $K e^u \in L^1(\mathbb{R})$ for a general class of positive, even and monotone-decreasing functions $K > 0$.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

26 extracted references · 25 canonical work pages

[3]

Breteaux, J

S. Breteaux, J. Faupin, and V. Grasselli , On the number of bound states for fractional S chr \"o dinger operators with critical and super-critical exponent , 2024

work page 2024
[1]

Ahrend and E

M. Ahrend and E. Lenzmann , Uniqueness for the nonlocal L iouville equation in R , J. Funct. Anal., 283 (2022), pp. Paper No. 109712, 29

work page 2022
[17]

F. P. Lanz , Ph.D. thesis . Work in preparation

work page
[2]

Beckner , Functionals for multilinear fractional embedding , Acta Math

W. Beckner , Functionals for multilinear fractional embedding , Acta Math. Sin. (Engl. Ser.), 31 (2015), pp. 1--28

work page 2015
[4]

J. A. Carrillo, M. G. Delgadino, J. Dolbeault, R. L. Frank, and F. Hoffmann , Reverse H ardy- L ittlewood- S obolev inequalities , J. Math. Pures Appl. (9), 132 (2019), pp. 133--165

work page 2019
[5]

Da Lio and L

F. Da Lio and L. Martinazzi , The nonlocal L iouville-type equation in R and conformal immersions of the disk with boundary singularities , Calc. Var. Partial Differential Equations, 56 (2017), pp. Paper No. 152, 31

work page 2017
[6]

DelaTorre, G

A. DelaTorre, G. Mancini, A. Pistoia, and L. Provenzano , Non-degeneracy of the bubble in a fractional and singular 1d liouville equation , 2025

work page 2025
[7]

Dou and M

J. Dou and M. Zhu , Reversed H ardy- L ittewood- S obolev inequality , Int. Math. Res. Not. IMRN, (2015), pp. 9696--9726

work page 2015

Show all 26 references

[8]

A. T. Duong and V. H. Nguyen , A L iouville-type theorem for fractional elliptic equation with exponential nonlinearity , Mediterr. J. Math., 19 (2022), pp. Paper No. 91, 16

2022
[9]

M. M. Fall , Entire s -harmonic functions are affine , Proc. Amer. Math. Soc., 144 (2016), pp. 2587--2592

2016
[10]

Farina , Stable solutions of - u=e^u on R^N , C

A. Farina , Stable solutions of - u=e^u on R^N , C. R. Math. Acad. Sci. Paris, 345 (2007), pp. 63--66

2007
[11]

Fazly, J

M. Fazly, J. Wei, and W. Yang , Classification of finite M orse index solutions of higher-order G elfand- L iouville equation , Discrete Contin. Dyn. Syst., 45 (2025), pp. 2001--2044

2025
[12]

R. L. Frank and E. Lenzmann , Uniqueness of non-linear ground states for fractional L aplacians in R , Acta Math., 210 (2013), pp. 261--318

2013
[13]

R. L. Frank, E. Lenzmann, and L. Silvestre , Uniqueness of radial solutions for the fractional L aplacian , Comm. Pure Appl. Math., 69 (2016), pp. 1671--1726

2016
[14]

G\'erard and E

P. G\'erard and E. Lenzmann , The C alogero- M oser derivative nonlinear S chr\"odinger equation , Comm. Pure Appl. Math., 77 (2024), pp. 4008--4062

2024
[15]

Gilbarg and N

D. Gilbarg and N. S. Trudinger , Elliptic partial differential equations of second order , Classics in Mathematics, Springer-Verlag, Berlin, 2001. Reprint of the 1998 edition

2001
[16]

Hyder and W

A. Hyder and W. Yang , Classification of stable solutions to a non-local G elfand- L iouville equation , Int. Math. Res. Not. IMRN, (2022), pp. 5219--5255

2022
[18]

F. P. Lanz and E. Lenzmann , Global existence and Blowup for _t u + (- ) ^s u = e^u in R . Work in preparation, 2026

2026
[19]

Li and M

Y. Li and M. Zhu , Uniqueness theorems through the method of moving spheres , Duke Math. J., 80 (1995), pp. 383--417

1995
[20]

E. H. Lieb and M. Loss , Analysis , vol. 14 of Graduate Studies in Mathematics, American Mathematical Society, Providence, RI, second ed., 2001

2001
[21]

Q. A. Ng \^o and V. H. Nguyen , Sharp reversed H ardy- L ittlewood- S obolev inequality on R ^n , Israel J. Math., 220 (2017), pp. 189--223

2017
[22]

Ros-Oton , Regularity for the fractional G elfand problem up to dimension 7 , J

X. Ros-Oton , Regularity for the fractional G elfand problem up to dimension 7 , J. Math. Anal. Appl., 419 (2014), pp. 10--19

2014
[23]

Ros-Oton and J

X. Ros-Oton and J. Serra , The extremal solution for the fractional L aplacian , Calc. Var. Partial Differential Equations, 50 (2014), pp. 723--750

2014
[24]

R. L. Schilling, R. Song, and Z. Vondra c ek , Bernstein functions , vol. 37 of De Gruyter Studies in Mathematics, Walter de Gruyter & Co., Berlin, second ed., 2012. Theory and applications

2012
[25]

J. I. Tello , Stability of steady states of the C auchy problem for the exponential reaction-diffusion equation , J. Math. Anal. Appl., 324 (2006), pp. 381--396

2006
[26]

Xu , Uniqueness and non-existence theorems for conformally invariant equations , J

X. Xu , Uniqueness and non-existence theorems for conformally invariant equations , J. Funct. Anal., 222 (2005), pp. 1--28

2005

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [3]

Breteaux, J

S. Breteaux, J. Faupin, and V. Grasselli , On the number of bound states for fractional S chr \"o dinger operators with critical and super-critical exponent , 2024

work page 2024

[2] [1]

Ahrend and E

M. Ahrend and E. Lenzmann , Uniqueness for the nonlocal L iouville equation in R , J. Funct. Anal., 283 (2022), pp. Paper No. 109712, 29

work page 2022

[3] [17]

F. P. Lanz , Ph.D. thesis . Work in preparation

work page

[4] [2]

Beckner , Functionals for multilinear fractional embedding , Acta Math

W. Beckner , Functionals for multilinear fractional embedding , Acta Math. Sin. (Engl. Ser.), 31 (2015), pp. 1--28

work page 2015

[5] [4]

J. A. Carrillo, M. G. Delgadino, J. Dolbeault, R. L. Frank, and F. Hoffmann , Reverse H ardy- L ittlewood- S obolev inequalities , J. Math. Pures Appl. (9), 132 (2019), pp. 133--165

work page 2019

[6] [5]

Da Lio and L

F. Da Lio and L. Martinazzi , The nonlocal L iouville-type equation in R and conformal immersions of the disk with boundary singularities , Calc. Var. Partial Differential Equations, 56 (2017), pp. Paper No. 152, 31

work page 2017

[7] [6]

DelaTorre, G

A. DelaTorre, G. Mancini, A. Pistoia, and L. Provenzano , Non-degeneracy of the bubble in a fractional and singular 1d liouville equation , 2025

work page 2025

[8] [7]

Dou and M

J. Dou and M. Zhu , Reversed H ardy- L ittewood- S obolev inequality , Int. Math. Res. Not. IMRN, (2015), pp. 9696--9726

work page 2015