A blow up solution of the Navier-Stokes equations with a critical force

Qi S. Zhang

REVIEW 3 major objections 4 minor 10 references

A blow up solution of the Navier-Stokes equations with a critical force

T0 review · 3 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 This paper constructs a forced Navier-Stokes solution that stays in the energy space but blows up at a finite time, with a forcing term at the critical scaling order -2.

desk verdict A genuine improvement over earlier forced blow-up constructions: an explicit critical-order force with zero initial data, but the proof skips the one estimate that puts the solution in the energy space. read the letter →

arxiv 2411.13896 v3 pith:UALXZYWP submitted 2024-11-21 math.AP

Qi S. Zhang This is my paper · ORCID

classification math.AP MSC 35Q3076N10

keywords Navier-Stokesequationsfinite-timeblow-upcriticalforcingenergyspaceStokeskernelfixedpointlog-subcriticalforce

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 aims to show that a forcing term at the critical scaling order $-2$ can drive a smooth solution of the Navier-Stokes equations to blow up in finite time, even when the initial velocity is zero and the solution remains in the energy space (finite kinetic energy, finite accumulated dissipation) throughout. The construction starts from a blow-up solution of a linear heat equation with the critical force, turns it into a divergence-free velocity via a curl-curl transform, and then adds a small correction obtained by a fixed-point argument. If the proof is right, the explicit force $-\delta \frac{e^{-|x|^2}}{(|x|^2+T-t)[1+|\ln(|x|^2+T-t)|]}(1,0,0)$ is enough to produce a singularity, suggesting that natural point-source forces with the same scaling could also cause singularity formation.

What carries the argument

The carrying object is the linear Stokes solution $Z=\operatorname{curl}\operatorname{curl}((-\Delta)^{-1}h)$, where $h$ solves the vector heat equation $\Delta h-\partial_t h=F$ with the critical force; the curl-curl operator makes $Z$ divergence-free and leaves a gradient error that is absorbed into the pressure. The correction $u$ is produced as the fixed point of an integral equation built from the Stokes kernel $K$, the fundamental solution of the linearized Stokes system, whose decay $|\nabla K|\le C/(|x-y|+\sqrt{t-s})^4$ makes the map a contraction in the weighted space with norm $\sup(1+|x|)^2|u|$. The pointwise estimates $|Z(x,t)|\le C\delta(1+|x|^2)^{-1}(1+|\ln(|x|^2+1-t)|)$ and a similar bound for $\nabla Z$ are what make the nonlinear terms in the equation for $u$ subcritical, so a small-data argument applies.

What would settle it

Take a fixed small $\delta>0$, evaluate the heat-kernel integral defining $h_1$ at $x=0$ as $t\to1^-$, and check that $|h_1(0,t)|$ diverges at least like $\ln\ln(1/(1-t))$; if it stays bounded, the linear solution $Z$ does not blow up. Then compute the fixed point $u$ of the integral equation and test whether $\int_0^1\int_{\mathbb{R}^3}|\nabla u|^2\,dx\,dt$ is finite; if that integral diverges, the claimed energy-space blow-up solution collapses.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1.2: there exists $\delta_0>0$ such that for every $\delta\in(0,\delta_0]$, the forced Navier-Stokes equations with zero initial data and force $F=-\delta \frac{e^{-|x|^2}}{(|x|^2+T-t)[1+|\ln(|x|^2+T-t)|]}(1,0,0)$ have a smooth solution $v$ in the energy space on $\mathbb{R}^3\times[0,T)$ that blows up at time $T$. The solution is written $v=Z+u$, where $Z$ is the explicitly defined linear Stokes solution produced by the heat-kernel formula and $u$ is the fixed point of a contraction in a weighted space. A stronger but less explicit version works in any open domain containing the origin with no-slip boundary conditions and also produces a force in the standard critical space $L^\infty_t L^{3/2}_x$. The proof uses the pressure to absorb a gradient error and checks that the nonlinear term $v\nabla v$ stays below the critical scaling, so the blow-up is driven by the linear, force-induced term.

Load-bearing premise

The proof stands on two pointwise decay estimates for the linear solution $Z$ and on the statement, not fully proven in the text, that the fixed-point correction $u$ has a square-integrable gradient on $\mathbb{R}^3\times[0,1)$; if either fails, $v=Z+u$ is not guaranteed to lie in the energy space and the blow-up solution is not established.

Editorial extensions

If this is right

For the explicit force with any sufficiently small amplitude $\delta$, the forced Navier-Stokes equations with zero initial data admit a smooth solution in the energy space that blows up at $T$.
The blow-up is driven by the linear Stokes solution $Z$, which diverges like $\ln\ln(1/(1-t))$ for the explicit force (or faster in the non-explicit variant), while the correction $u$ stays bounded.
The same construction works in any open domain containing the origin with no-slip boundary conditions when the force's principal part is the explicit one, and it also yields forces in the critical space $L^\infty_t L^{3/2}_x$.
Because the force has scaling order $-2$, matching point-source forces, the result shows that a critical (not supercritical) driving term can produce a finite-time singularity.

Reading between the lines

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

If the contraction proof is stable, the same $v=Z+u$ splitting should yield blow-up for a family of forces obtained by adding small, smooth, divergence-free perturbations to the explicit $F$, as long as they stay at or below the critical scale.
The paper leaves the sign of $F\cdot v$ uncontrolled; a natural next step would be to construct a version with $F\cdot v\ge0$, which would make the force plausibly 'slowing' rather than 'pushing' the flow and strengthen the physical interpretation.
The result is a step toward singular solutions of the unforced equations only if the pressure and force can eventually be removed; nothing in this paper shows that, so the step remains incomplete.
A numerical check of the predicted $\ln\ln(1/(1-t))$ blow-up rate at $x=0$ could test whether the linear mechanism is the one actually operating for finite $\delta$.

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 real step past the author's own earlier supercritical example, and Theorem 1.2 is the part that matters—an explicit force of critical order −2, fixed before the solution is constructed, giving a smooth blow-up in the energy space with zero initial data. I think the paper deserves refereeing. It also has one genuinely under-verified step that a referee should push on.

What's new: the force scaling is now critical rather than supercritical, and the log-subcritical variant in Theorem 1.1(b) plus the explicit L∞_t L^{3/2}_x force in Theorem 1.2 are genuine improvements over [9] and Scheffer's "bizarre function". The construction itself is straightforward—solve the heat equation with the forcing, turn the solution into a divergence-free field via curl curl, absorb the error into pressure—but the details matter and most of them are checked. The blow-up proof via radial symmetry is clean, and the energy bound for the linear solution through Riesz transforms is solid. The fixed point for u in Theorem 1.2 is done with care; the bounds on T4, T3, T2, T1 are explicit and the estimates (2.39)–(2.44) check out.

Where I'd be cautious: the step after (2.48) is the load-bearing one for the energy-space claim. The contraction gives u only in the weighted sup-norm space X. To conclude v = Z + u is in the energy space you need ∇u ∈ L^2_t L^2_x, and the paper asserts this from "standard parabolic singular integral theory" without showing that the four products u⊗u, Z⊗u, u⊗Z, Z⊗Z are in L^2_{t,x}. The decay bounds (2.27) and |u| ≤ η/(1+|x|^2) make that plausible, and I don't think it's wrong, but it's exactly the kind of estimate that should be written down, especially since the later integration by parts in (2.49) already assumes ∇u is integrable. A referee should ask for a lemma here.

Part (b) of Theorem 1.1 is also summarized rather than proved; the phrase "step by step with only small changes" is doing a lot of work. I believe it follows, but it should be laid out.

The circularity concern is mostly handled: Theorem 1.1 is reverse-engineered by design, and the paper says so. Theorem 1.2 is the honest version, with the force fixed first. The remaining question is only the technical check above.

Who is this for? People working on forced Navier-Stokes blow-up, Scheffer's program, or critical regularity thresholds. It doesn't speak directly to the unforced regularity problem, and the authors are appropriately careful about that.

Recommendation: send it out. It's a constructive example with an explicit force, and the central idea is sound. The referee should require a complete proof of the gradient estimate for u and a fuller treatment of Theorem 1.1(b).

Referee Report

3 major / 4 minor

Summary. The paper constructs finite-time blow-up solutions of the forced, incompressible Navier-Stokes equations in three dimensions. Theorem 1.1(a) gives, on any domain containing the origin, a smooth compactly supported solution in the energy space whose force has principal part -e^{-|x|^2}/(|x|^2+T-t)(1,0,0). Theorem 1.1(b) gives a smooth solution on R^3 with force in L^\infty_t L^{3/2}_x whose principal part contains an additional logarithmic factor. Theorem 1.2 gives, for all sufficiently small \delta>0, a smooth energy-space solution with zero initial data and the fully explicit force F=-\delta e^{-|x|^2}/((|x|^2+T-t)(1+|\ln(|x|^2+T-t)|))(1,0,0), which blows up at time T. The method solves a scalar forced heat equation, applies the transformation v=curl(curl((-\Delta)^{-1}h)), absorbs gradient errors into the pressure, and for Theorem 1.2 writes v=Z+u where Z is the linear Stokes solution and u is obtained by a contraction mapping in a weighted sup-norm space.

Significance. If the gaps identified below are repaired, this would be a significant advance: the forcing in Theorems 1.1(b) and 1.2 is at or below the critical scaling order -2, unlike earlier supercritical constructions, and Theorem 1.2 provides an explicit force with zero initial data and finite-time blow-up in the energy space. The paper is also careful to avoid the trivial 'define F after v' objection by prescribing F before solving for the perturbation in Theorem 1.2. The proof contains several clean and useful estimates, especially the Riesz-transform argument showing blow-up of v at the origin and the weighted contraction argument. However, the manuscript as it stands has a load-bearing gap in the justification of the energy-space membership of the perturbation u in Theorem 1.2, and part (b) of Theorem 1.1 is not proved with sufficient detail.

major comments (3)

[Section 2, proof of Theorem 1.2, after Eq. (2.48)] The statement 'Due to the boundedness of u and ln type singularity of Z, the standard parabolic singular integral theory tells us |\nabla u| \in L^2_t L^2_x' is asserted without proof. The fixed point is obtained in the sup-norm space X of (2.32), which gives no gradient information. To apply the L^2-boundedness of second derivatives of the Stokes kernel, one must first show that each product u_i u, Z_i u, u_i Z, Z_i Z lies in L^2_t L^2_x. For Z_i Z this requires checking \int_0^1\int |Z|^4 dxdt<\infty; this is plausible from (2.27) and the fact that (1+|x|^2)^{-4} is integrable in R^3, but no computation is given. For the mixed products one needs \int_0^1\int |Z|^2|u|^2 dxdt<\infty, which likewise is not written out. Moreover, the integration by parts that yields (2.49) already presupposes \nabla u \in L^2, so if the singular-integral step is not justified first, the regularity bootstrap is circular. Since membership of v in the energy space is an explicit conclusion of Theorem 1.2, this omitted verification is a load-bearing gap and must be supplied with explicit estimates.
[Section 2, proof of Theorem 1.1(b), around Eq. (2.23)] The proof of part (b) is summarized as 'The rest of the proof follows that of part (a) step by step with only small changes.' This is insufficient for the theorem's central claim that F=f-v\nabla v belongs to L^\infty_t L^{3/2}_x. In particular, one needs explicit bounds showing that both the heat forcing f in (2.20) and the nonlinear term v\nabla v, estimated through the log-log analogues of (2.18) and (2.19), have finite L^{3/2}_x norm uniformly in t. The sentence 'It is now easy to see the forcing term F=f-v\nabla v is in the space L^\infty_t L^{3/2}_x' is too terse for a theorem-level assertion; a short verification should be included.
[Section 2, proof of Theorem 1.2, Eqs. (2.33) and (2.45)] The contraction argument itself is mostly sound, but the notation for the Stokes kernel convolution is imprecise: in (2.33) the expression '\partial_i K u_i u(y,s)' should be written with parentheses, e.g. '\partial_{y_i}K(x,t;y,s)\, u_i(y,s)u(y,s)', and similarly in (2.45). More substantively, the proof of the containment M(B(0,\eta))\subset B(0,\eta) uses the estimate (2.42) for T_4 that contains a factor C\delta^2; the subsequent choice of \eta and \delta is legitimate, but the dependence of the constants on the logarithmic factors should be tracked to ensure that the smallness condition is uniform in \delta and \eta. This is a presentation issue rather than a logical flaw, but it would help the reader to see the explicit constants.

minor comments (4)

[Section 2, proof of Theorem 1.2, first paragraph] The sentence 'Just like Theorem 1.2, Z is in the energy space and it blows up at t=1' should refer to Theorem 1.1, not Theorem 1.2.
[Equation (2.47)] The denominator in the last inequality is written as '|1+|x|^2'; it should be '1+|x|^2'.
[Throughout] Minor language issues: 'scales logarithmic worse than -1' should be 'scales logarithmically worse than -1'; '0 initial value' should be 'zero initial value'; 'through out' should be 'throughout'.
[Theorem 1.1(a) statement] The phrase 'smooth, compactly supported solution v' is potentially confusing because v is constructed with a spatial cutoff \varphi and is compactly supported for each time, but it is not compactly supported in time on [0,1). The statement could be clarified.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the Theorem 1.2 force is fixed before the solution is built, and the Theorem 1.1 residual-force construction is an existence proof, not a fitted prediction.

full rationale

No circularity is present. In Theorem 1.1 the force is defined after v, namely F = fφ − v∇v + f_b in (2.11), which is a standard existence construction of a force with the desired critical-order principal term, not a prediction from an independently prescribed force. The paper explicitly distinguishes this from the trivial 'define F as the residual' trick and then addresses the remaining concern in Theorem 1.2, where the force is fixed first: 'the Navier Stokes equations (1.1) with the L∞_t L^{3/2}_x force F = ... and 0 initial value have a smooth solution v in the energy space in R^3 × [0, T), which blows up at time T.' The proof solves the linear Stokes problem with this fixed F to obtain Z, estimates Z via (2.27)-(2.28), and then runs a contraction mapping for u in the weighted norm X; the bounds (2.36)-(2.44) for the four convolution terms are proved in the paper itself rather than imported from prior work. The one load-bearing analytic assertion that is not fully demonstrated is the L^2_t L^2_x bound on ∇u after (2.48); this is an omitted estimate or correctness risk, not a definitional reduction, and it does not make the theorem's conclusion equal to its input by construction. The citations to [10] for the Stokes convolution are invoked for well-definedness in broader function classes, but the contraction estimates needed here are re-proved in the text, so the self-citations are not load-bearing. The paper's own caveats (lower-order force terms are not explicit, and the sign of F·v is not controlled) are limitations about physicality, not indicators of circular reasoning.

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

No fitted parameters appear in the construction; the smallness parameter δ in Theorem 1.2 is an existence threshold, not a fitted constant. The central claim rests on standard heat kernel and Stokes kernel estimates, Riesz transform L^2 boundedness, and parabolic singular integral theory. No new physical or mathematical entities are postulated.

assumptions (3)

standard math Stokes kernel pointwise bounds (2.34): |K| ≤ C/(|x-y|+√(t-s))^3 and |∇K| ≤ C/(|x-y|+√(t-s))^4, cited from Solonnikov [8].
Used throughout the contraction mapping estimates in Theorem 1.2; the constants are required to be uniform up to t=1.
standard math Riesz transforms are bounded on L^2; the Leray projection v = curl curl((−Δ)^{-1}h) and its gradient are controlled by h and ∇h in L^2.
Used in the energy estimates (2.13)-(2.16) and to transfer blow-up of h to blow-up of v.
standard math Parabolic singular integral operators with the Stokes kernel map L^2_t L^2_x to themselves.
Invoked after equation (2.48) to conclude that the fixed point u has ∇u in L^2_t L^2_x; a full proof is not included in the paper.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A blow up solution of the Navier-Stokes equations with a critical force." pith.science (2026). https://pith.science/paper/UALXZYWP

@misc{pith2026241113896,
  author       = {Pith},
  title        = {Pith review of: A blow up solution of the Navier-Stokes equations with a critical force},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UALXZYWP}},
  note         = {Machine review of arXiv:2411.13896}
}

abstract

A forced solution $v$ of the Navier-Stokes equation in any open domain with no slip boundary condition is constructed. The scaling factor of the forcing term is the critical order $-2$. The velocity, which is smooth until its final blow up moment, is in the energy space through out. Since most physical forces from a point source in nature are regarded as order $-2$, such as Coulomb force, Yukawa force, this result indicates possible singularity formation under these kind of forces. The result even holds for some log subcritical forces or some forces in the standard critical space $L^\infty_t L^{3/2}_x$, including the explicit force: $F=- \delta \frac{e^{-|x|^2}}{(|x|^2 + T-t) \,[1+ | \ln (|x|^2 + T-t)|]} (1, 0, 0) $ for any small $\delta>0$. The result can also be considered as a step in Scheffer's plan.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

10 extracted references · 9 canonical work pages

[9]

Zhang, Qi S., A blow up solution of the Navier-Stokes equations with a supe r critical forcing term , arXiv:2311.12306. 16 Q. S. ZHANG

work page Pith review arXiv
[1]

Colombo, Non-uniqueness of Leray solutions of the forced Navier- Stokes equations

D.Albritton, E.Bru´ e, and M. Colombo, Non-uniqueness of Leray solutions of the forced Navier- Stokes equations. Ann. of Math. (2), 196(1):415-455, 2022

work page 2022
[2]

Nonuniqueness of weak solutions to the Navier-Stokes equa- tion

Tristan Buckmaster and Vlad Vicol. Nonuniqueness of weak solutions to the Navier-Stokes equa- tion. Ann. of Math. (2), 189(1):101-144, 2019

work page 2019
[3]

Hugo Beiro da Veiga and Jiaqi Yang, A note on the development of singularities on solutions to the Navier-Stokes equations under super critical forcing t erms, arXiv:2411.10823

work page Pith review arXiv
[4]

and Zheng, F.: Finite time blow-up for the hypodissipative Navier Stokes equations with a force in L1 t C1,ǫ x ∩ L∞ t L2 x arXiv preprint arXiv:2407.06776, 2024

Cordoba, D., Martinez-Zoroa, L. and Zheng, F.: Finite time blow-up for the hypodissipative Navier Stokes equations with a force in L1 t C1,ǫ x ∩ L∞ t L2 x arXiv preprint arXiv:2407.06776, 2024

arXiv 2024
[5]

Leray, Sur le mouvement d’un liquide visqueux emplissant l’espace

J. Leray, Sur le mouvement d’un liquide visqueux emplissant l’espace . Acta Math. 63 (1934), 193–248

work page 1934
[6]

Zhang, Na Zhao , Finite speed axially symmetric Navier-Stokes ﬂows passing a cone , J

Zijin Li, Xinghong Pan, Xin Yang, Chulan Zeng, Qi S. Zhang, Na Zhao , Finite speed axially symmetric Navier-Stokes ﬂows passing a cone , J. Functional Analysis, Volume 286, Issue 10, 15 May 2024, 116pp

work page 2024
[7]

Scheﬀer, A Solution to the Navier-Stokes Inequality with an Internal Singularity , Commun

V. Scheﬀer, A Solution to the Navier-Stokes Inequality with an Internal Singularity , Commun. Math. Phys. 101, 47-85 (1985)

work page 1985

Show all 10 references

[8]

V. A. Solonnikov, Estimates Solving Nonstationary Linearized Systems of Nav ier-Stokes Equa- tions, Trudy Mat. Inst. Steklov, Vol. 70, 1964. pp. 213-317

1964
[10]

Zhang, Qi S., Global solutions of Navier-Stokes equations with large L2 norms in a new function space, Advances in Diﬀerential Equations, Vol. 9, No. 5-6, 2004, pp 587- 624 Department of mathematics, University of California, Rive rside, CA 92521, USA Email address : qizhang@m...

2004

Pith tools

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

[1] [9]

Zhang, Qi S., A blow up solution of the Navier-Stokes equations with a supe r critical forcing term , arXiv:2311.12306. 16 Q. S. ZHANG

work page Pith review arXiv

[2] [1]

Colombo, Non-uniqueness of Leray solutions of the forced Navier- Stokes equations

D.Albritton, E.Bru´ e, and M. Colombo, Non-uniqueness of Leray solutions of the forced Navier- Stokes equations. Ann. of Math. (2), 196(1):415-455, 2022

work page 2022

[3] [2]

Nonuniqueness of weak solutions to the Navier-Stokes equa- tion

Tristan Buckmaster and Vlad Vicol. Nonuniqueness of weak solutions to the Navier-Stokes equa- tion. Ann. of Math. (2), 189(1):101-144, 2019

work page 2019

[4] [3]

Hugo Beiro da Veiga and Jiaqi Yang, A note on the development of singularities on solutions to the Navier-Stokes equations under super critical forcing t erms, arXiv:2411.10823

work page Pith review arXiv

[5] [4]

and Zheng, F.: Finite time blow-up for the hypodissipative Navier Stokes equations with a force in L1 t C1,ǫ x ∩ L∞ t L2 x arXiv preprint arXiv:2407.06776, 2024

Cordoba, D., Martinez-Zoroa, L. and Zheng, F.: Finite time blow-up for the hypodissipative Navier Stokes equations with a force in L1 t C1,ǫ x ∩ L∞ t L2 x arXiv preprint arXiv:2407.06776, 2024

arXiv 2024

[6] [5]

Leray, Sur le mouvement d’un liquide visqueux emplissant l’espace

J. Leray, Sur le mouvement d’un liquide visqueux emplissant l’espace . Acta Math. 63 (1934), 193–248

work page 1934

[7] [6]

Zhang, Na Zhao , Finite speed axially symmetric Navier-Stokes ﬂows passing a cone , J

Zijin Li, Xinghong Pan, Xin Yang, Chulan Zeng, Qi S. Zhang, Na Zhao , Finite speed axially symmetric Navier-Stokes ﬂows passing a cone , J. Functional Analysis, Volume 286, Issue 10, 15 May 2024, 116pp

work page 2024

[8] [7]

Scheﬀer, A Solution to the Navier-Stokes Inequality with an Internal Singularity , Commun

V. Scheﬀer, A Solution to the Navier-Stokes Inequality with an Internal Singularity , Commun. Math. Phys. 101, 47-85 (1985)

work page 1985