Singularities of the Lagrangian mean curvature flow at the critical Lagrangian phase
Arunima Bhattacharya, Ravi Shankar, Jeremy Wall, and Diego Yepez · 2025
Read Singularities of the Lagrangian mean curvature flow at the critical Lagrangian phase by Arunima Bhattacharya, Ravi Shankar, Jeremy Wall, and Diego Yepez (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-22741.txt.
Introduction
Surfaces that move by their own curvature can form sharp singularities. Arunima Bhattacharya leads a team with three co-authors. In 2025 they study a Lagrangian version of this flow. The key setting is the critical phase case. Phase here is an angle like quantity in the model.
The team proves interior estimates near such singularities. Their gradient bounds need some structural conditions. They also build rough singular solutions. Those examples show the conditions cannot be dropped. A new proof method raises a key operator to a power. That makes the operator concave, which helps the estimates. Results extend to a wider class of equations.
Catalog record
| CREATOR | Arunima Bhattacharya, Ravi Shankar, Jeremy Wall, and Diego Yepez |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2510-22741
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Singularities of the Lagrangian mean curvature flow at the critical Lagrangian phase Arunima Bhattacharya, Ravi Shankar, Jeremy Wall, and Diego Yepez Abstract . We establish interior estimates for singularities of the Lagrangian mean curvature flow when the Lagrangian phase is critical, i.e., | Θ | ≥ ( n − 2 ) π 2 |\Theta|\geq(n-2)\tfrac{\pi}{2} , and extend our results to the broader class of Lagrangian mean curvature type equations. Our gradient estimates require certain structural conditions, and we construct C α C^{\alpha} singular viscosity solutions to show that criticality of the phase is necessary, and that these conditions cannot be removed in dimension one. We also introduce a new method for proving C 2 , α C^{2,\alpha} estimates by exponentiating the arctangent operator into a concave one when | Θ | ≥ ( n − 2 ) π 2 |\Theta|\geq(n-2)\tfrac{\pi}{2} and n > 2 n>2 . 1. Introduction We say that u solves a Lagrangian mean curvature type equation in the complex Euclidean space when (1.1) \sum_{i=1}^{n}\arctan\lambda_{i}=\Theta(x,u,Du)\in(-n\frac{\pi}{2},n\frac{\pi}{2}), where \lambda_{i} ’s are the eigenvalues of the Hessian of u . The quantity \Theta , determined by (1.2) \Theta=\sum_{i=1}^{n}\arctan\lambda_{i}, is called the Lagrangian phase or angle of the Lagrangian submanifold L=(x,Du(x))\subset\mathbb{C}^{n} . By Harvey-Lawson [ HL82 ] , the Lagrangian phase determines the mean curvature vector of L : (1.3) \vec{H}=J\nabla_{g}\Theta, where \nabla_{g} is the gradient operator for the induced metric g_{ij}=\delta_{ij}+u_{ik}\delta^{kl}u_{lk} on L , and J is the \frac{\pi}{2} rotation matrix in \mathbb{C}^{n} . When the Lagrangian phase is constant, i.e., \Theta=c , one gets the special Lagrangian equation . In this case, H=0 , and L is a volume-minimizing submanifold in \mathbb{C}^{n} . A dual form of the special Lagrangian equation is the Monge-Ampére equation \det D^{2}u=c . This is the potential equation for special Lagrangian submanifolds in (\mathbb{R}^{n}\times\mathbb{R}^{n},dxdy) as interpreted by Hitchin in [ Hit97 ] . We say that u solves the potential equation for prescribed Lagrangian mean curvature when \sum_{i=1}^{n}\arctan\lambda_{i}=\Theta(x). A family of Lagrangian submanifolds X(x,t):\mathbb{R}^{n}\times\mathbb{R}\to\mathbb{C}^{n} evolves by Lagrangian mean curvature flow (LMCF) if it solves (X_{t})^{\bot}=\Delta_{g}X=\vec{H}. After a change of coordinates, one can locally write X(x,t)=(x,Du(x,t)) , such that \Delta_{g}X=(J\bar{\nabla}\Theta(x,t))^{\bot} . This means a local potential u(x,t) evolves by the parabolic equation \displaystyle u_{t} \displaystyle=\sum_{i=1}^{n}\arctan\lambda_{i}, \displaystyle u(x,0):=u(x). Symmetry reductions of the Lagrangian mean curvature flow reduce the above local parabolic representation to an elliptic equation for u(x) , which models singularities of the mean curvature flow (see Chau-Chen-He [ CCH12 ] ). If u(x) solves (1.4) \displaystyle\sum_{i=1}^{n}\arctan\lambda_{i}=s_{1}+s_{2}(x\cdot Du(x)-2u(x)), then X(x,t)=\sqrt{1-2s_{2}t}\,(x,Du(x)) is a shrinker or expander solution of the LMCF, if s_{2}>0 or s_{2}<0 respectively. If u(x) solves (1.5) \displaystyle\sum_{i=1}^{n}\arctan\lambda_{i}=\gamma_{1}+\gamma_{2}\cdot x+\gamma_{3}\cdot Du(x), then X(x,t)=(x,Du(x))+t(-\gamma_{3},\gamma_{2}) is a translator solution of the LMCF. If u(x) solves (1.6) \displaystyle\sum_{i=1}^{n}\arctan\lambda_{i}=r_{1}+\frac{r_{2}}{2}(|x|^{2}+|Du(x)|^{2}), then X(x,t)=\exp(r_{2}tJ)(x,Du(x)) is a rotator solution of the LMCF. Notation Before we present our main results, we clarify some terminology. I. By B_{R} we denote a ball of radius R centered at the origin. II. We denote the oscillation of u in B_{R} by \mathrm{osc}_{B_{R}}(u) . III. Let \Gamma_{R}=B_{R}\times u(B_{R})\times Du(B_{R})\subset B_{R}\times\mathbb{R}\times\mathbb{R}^{n} . Let \nu_{1},\nu_{2} be constants such that for \Theta(x,z,p) , we assume the following structure conditions (1.7) \displaystyle|\Theta_{x}|,|\Theta_{z}|,|\Theta_{p}| \displaystyle\leq\nu_{1}(\mathrm{osc}_{B_{R}}u,\|Du\|_{L^{\infty}(B_{R})}), \displaystyle|\Theta_{xx}|,|\Theta_{xz}|,|\Theta_{xp}|,|\Theta_{zz}|,|\Theta_{zp}| \displaystyle\leq\nu_{2}(\mathrm{osc}_{B_{R}}u,\|Du\|_{L^{\infty}(B_{R})}), for all (x,z,p)\in\Gamma_{R} . In the above partial derivatives, the variables x,z,p are treated as independent of each other. Observe that this indicates that the above partial derivatives do not have any D^{2}u or D^{3}u terms. Our main results are the following. Theorem 1.1 (Hessian Estimates) . If u is a smooth solution of ( 1.1 ) on B_{R}\subset\mathbb{R}^{n} , with n\geq 3 and |\Theta|\geq(n-2)\frac{\pi}{2} where \Theta(x,z,p)\in C^{2}(\Gamma_{R}) satisfies ( 1.7 ) and is partially convex in p , then we get |D^{2}u(0)|\leq C_{1}\exp{[C_{2}\|Du\|_{L^{\infty}(B_{R})}^{2n+3}/R^{2n+3}]} where C_{1} and C_{2} are positive constants depending on \nu_{1},\nu_{2} , and n . Remark 1.1 . From the singular solutions constructed in [ BS24 , (1.13)] , it is clear that the Hessian estimates stated in Theorem 1.1 fail to hold in the absence of partial convexity of \Theta with respect to the gradient variable Du . Corollary 1.1 . If u is a smooth solution of any of these equations: ( 1.4 ), ( 1.5 ), and ( 1.6 ) on B_{R}\subset\mathbb{R}^{n} where n\geq 3 and |\Theta|\geq(n-2)\frac{\pi}{2} , then we obtain |D^{2}u(0)|\leq C_{1}\exp{[C_{2}\|Du\|_{L^{\infty}(B_{R})}^{2n+5}/R^{2n+5}]} where C_{1} and C_{2} are positive constants depending on n and the following: s_{2} for ( 1.4 ), \gamma_{2},\gamma_{3} for ( 1.5 ), and r_{2} for ( 1.6 ). Remark 1.2 . In the case of ( 1.5 ), we get a better estimate |D^{2}u(0)|\leq C_{1}\exp{[C_{2}\|Du\|_{L^{\infty}(B_{R})}^{2n+3}/R^{2n+3}]}. Next, our gradient estimates are as follows. Theorem 1.2 (Gradient Estimates) . Let u be a smooth solution of ( 1.1 ) on B_{1}\subset\mathbb{R}^{n} with |\Theta|\geq(n-2)\frac{\pi}{2} . Suppose that \Theta(x,z,p) satisfies the following conditions: (a) \Theta(x,z,p) is uniformly C^{2} in x , i.e., |\Theta_{xx}|\leq\|\Theta_{xx}\|_{L^{\infty}(\Gamma_{1})} independently of u and Du , and that (1.8) \lvert D_{x}\Theta(x,z,p)\rvert\leq C\Bigl(\Theta(x,z,p)-\tfrac{(n-2)\pi}{2}\Bigr)^{1/2}. (b) \Theta_{z} is non-negative: (1.9) D_{z}\Theta(x,z,p)\;=\;\Theta_{z}(x,z,p)\;\geq 0. (c) \Theta has superlinear decay o(1/\lvert p\rvert) as p\to\infty and \Theta\to(n-2)\pi/2 , more quantitatively (1.10) \lvert p\rvert\,\lvert D_{p}\Theta(x,z,p)\rvert\;\leq\;C\Bigl(\Theta(x,z,p)-\tfrac{(n-2)\pi}{2}\Bigr). Then u satisfies the following estimate: |Du(0)|\leq C(n,\|\Theta_{xx}\|_{L^{\infty}(\Gamma_{1})})\left(1+(\text{osc}_{B_{1}}u)^{2}\right). Remark 1.3 . In Section 4 , we construct singular C^{\alpha} viscosity solutions to show that none of the conditions (a), (b), and (c) can be removed in dimension one. For solutions of the special Lagrangian equation with critical and supercritical phase |\Theta|\geq(n-2)\frac{\pi}{2} , Hessian estimates were obtained by Warren-Yuan [ WY09 , WY10 ] , Wang-Yuan [ WY14 ] , Li [ Li19 ] via a compactness approach, Shankar [ Sha24 ] via a doubling approach, and Zhou [ Zho22 ] for equations requiring Hessian constraints which generalize criticality. The singular C^{1,\alpha} solutions to ( 1.2 ) constructed by Nadirashvili-Vlăduţ [ NV10 ] and Wang-Yuan [ WY13 ] show that interior regularity is not possible for subcritical phases |\Theta|<(n-2)\frac{\pi}{2} , without an additional convexity condition, as shown in Bao-Chen [ BC03 ] , Chen-Warren-Yuan [ CWY09 ] , and Chen-Shankar-Yuan [ CSY23 ] , and that the Dirichlet problem is not classically solvable for arbitrary smooth boundary data. Recently, viscosity solutions to ( 1.2 ) that are Lipschitz but not C^{1} were constructed by Mooney-Savin [ MS23 ] . For solutions of the Lagrangian mean curvature equation, Hessian estimates for convex smooth solutions with C^{1,1} phase were obtained by Warren in [ War08 , Theorem 8] . For C^{1,1} critical and supercritical phase, interior Hessian and gradient estimates were established by Bhattacharya [ Bha21 , Bha22 ] and Bhattacharya-Mooney-Shankar [ BMS22 ] (for C^{2} phase), respectively. See also Lu [ Lu23 ] . Interior Hessian estimates for supercritical C^{0,1} phase were derived by Zhou [ Zho23 ] . Recently, interior Hessian estimates for critical and supercritical C^{0,1} phases were proven by Ding [ Din24 ] . For convex viscosity solutions, interior regularity was established for C^{2} phase by Bhattacharya-Shankar in [ BS23 ] and optimal regularity for the broader class of equations, given by ( 1.1 ) was derived in [ BS24 ] . If \Theta is merely in C^{\alpha} and supercritical, |\Theta|\geq(n-2)\frac{\pi}{2}+\delta , counterexamples to Hessian estimates exist as shown in [ Bha24 ] . For more work on the special Lagrangian and Lagrangian mean curvature type equations, we refer the reader to [ CW24 , WHB24 , QZ24 ,