Abstract
The Special Lagrangian Potential Equation for a function u on a domain \(\Omega \subset {{\mathbb {R}}}^n\) is given by \({\mathrm{tr}}\{\arctan (D^2 \,u) \} = \theta \) for a contant \(\theta \in (-n {\pi \over 2}, n {\pi \over 2})\). For \(C^2\) solutions the graph of Du in \(\Omega \times {{\mathbb {R}}}^n\) is a special Lagrangian submanfold. Much has been understood about the Dirichlet problem for this equation, but the existence result relies on explicitly computing the associated boundary conditions (or, otherwise said, computing the pseudo-convexity for the associated potential theory). This is done in this paper, and the answer is interesting. The result carries over to many related equations—for example, those obtained by taking \(\sum _k \arctan \, \lambda _k^{{\mathfrak {g}}}= \theta \) where \({{\mathfrak {g}}}: {\mathrm{Sym}}^2({{\mathbb {R}}}^n)\rightarrow {{\mathbb {R}}}\) is a Gårding-Dirichlet polynomial which is hyperbolic with respect to the identity. A particular example of this is the deformed Hermitian–Yang–Mills equation which appears in mirror symmetry. Another example is \(\sum _j \arctan \kappa _j = \theta \) where \(\kappa _1,\ldots , \kappa _n\) are the principal curvatures of the graph of u in \(\Omega \times {{\mathbb {R}}}\). We also discuss the inhomogeneous Dirichlet Problem
where \(\psi : {\overline{\Omega }}\rightarrow (-n {\pi \over 2}, n {\pi \over 2})\). This equation has the feature that the pull-back of \(\psi \) to the Lagrangian submanifold \(L\equiv {\mathrm{graph}}(Du)\) is the phase function \(\theta \) of the tangent spaces of L. On L it satisfies the equation \(\nabla \psi = -JH\) where H is the mean curvature vector field of L.
Similar content being viewed by others
References
Brendle, S., Warren, M.: A boundary value problem for minimal Lagrangian graphs. J. Differ. Geom. 84, 267–287 (2010)
Caffarelli, L., Nirenberg, L., Spruck, J.: The Dirichlet problem for nonlinear second order elliptic equations. I: Monge-Ampère equation. Commun. Pure Appl. Math. 37, 369–402 (1984)
Chen, G.: On J-equation, ArXiv preprints (2019). arXiv:1905.10222
Chen, G.: Supercritical deformed Hemitian–Yang–Mills equation. ArXiv preprints (2020). arXiv:2005.12202
Chen, J., Shankar, R., Yuan, Y.: Regularity for convex viscosity solutions of the special Lagrangian equation. arXiv:1911.05452
Chen, X.-X., Donaldson, S., Sun, S.: Kähler–Einstein metrics on Fano manifolds. I. Approximation of metrics with cone singularities. J. Am. Math. Soc. 28(1), 183–197 (2015)
Chen, X.-X., Donaldson, S., Sun, S.: Kähler–Einstein metrics on Fano manifolds. II. Limits with cone angle less than 2. J. Am. Math. Soc. 28(1), 199–234 (2015)
Chen, X.-X., Donaldson, S., Sun, S.: Kähler–Einstein metrics on Fano manifolds. III. Limits as cone angle approaches 2 and completion of the main proof. J. Am. Math. Soc. 28(1), 235–278 (2015)
Cirant, M., Payne, K.: Comparison principles for viscosity solutions of elliptic branches of fully nonlinear equations independent of the gradient. arXiv:2001.09658
Collins, T., Jacob, A., Yau, S.-T.: (1, 1) forms with specified Lagrangian phase: a priori estimates and algebraic obstructions. arXiv:1508.01934
Collins, T., Picard, S., Wu, X.: Concavity of the Lagrangian phase operator and applications. Calc. Var. Partial Differ. Equ. 56(4), Art. 89 (2017). arXiv:1607.07194
Collins, T.C., Xie, D., Yau, S.-T.: The deformed Hermitian–Yang–Mills equation in geometry and physics. arXiv:1712.00893
Collins, T.C., Shi, Y.: Stability and the deformed Hermitian–Yang–Mills equation. arXiv:2004.04831
Collins, T.C., Yau, S.T.: Moment maps, nonlinear pde, and stability in mirror symmetry. arXiv:1811.04824
Clarke, A., Smith, G.: The Perron method and the non-linear Plateau problem. Geom. Dedicata 163(1), 159–165 (2013)
Dinew, S., Do, H-S., Tô, T.D.: A viscosity approach to the Dirichlet problem for degenerate complex Hessian type equations. arXiv:1712.08572
Dellatorre, M.: The degenerate special Lagrangian equation on Riemannian manifolds. Int. Math. Res. Not. (to appear)
Donaldson, S.K.: Moment Maps and diffeomorphisms, Sir Michael Atiyah: a great mathematician of the twentieth century. Asian J. Math. 3(1), 1–15 (1999)
Fu, L.: An analogue of Bernstein’s theorem. Houst. J. Math. 24, 415–419 (1998)
Gårding, L.: An inequality for hyperbolic polynomials. J. Math. Mech. 8(2), 957–965 (1959)
Harvey, F.Reese, Lawson, H.Blaine: Calibrated geometries. Acta Math. 148, 47–157 (1982)
Harvey, F.Reese, Lawson, H.Blaine: Dirichlet duality and the non-linear Dirichlet problem. Commun. Pure Appl. Math. 62, 396–443 (2009)
Harvey, F.Reese, Lawson, H.Blaine: Dirichlet duality and the non-linear Dirichlet problem on Riemannian manifolds. J. Differ. Geom. 88(3), 395–482 (2011). arXiv:0907.1981
Harvey, F.R., Lawson, H.B.: Hyperbolic polynomials and the Dirichlet problem. arXiv:0912.5220
Harvey, F.Reese, Lawson, H.Blaine: Gårding’s theory of hyperbolic polynomials. Commun. Pure Appl. Math. 66(7), 1102–1128 (2013)
Harvey, F.Reese, Lawson, H.Blaine: The restriction theorem for fully nonlinear subequations. Ann. Inst. Fourier 64(1), 217–265 (2014). arXiv:1101.4850
Harvey, F.Reese, Lawson, H.Blaine: Lagrangian potential theory and a Lagrangian equation of Monge–Ampère type. In: Cao, H.-D., Li, J., Schoen, R., Yau, S.-T. (eds.) Surveys in Differential Geometry, vol. 22, pp. 217–257. International Press, Somerville (2018). arXiv:1712.03525
Harvey, F.R., Lawson, H.B.: The inhomogeneous Dirichlet problem for natural operators on manifolds. arXiv:1805.11121
Hitchin, N.J.: The moduli space of special Lagrangian submanifolds. Ann. Scuola Norm. Sup. Pisa Cl. Sci. 25(3–4), 503–515 (1997)
Jacob, A.: Weak geodesics for the deformed Hermitian–Yang–Mills equation. arXiv:1906.07128
Jacob, A., Yau, S.-T.: A special Lagrangian type equation for holomorphic line bundles. Math. Ann. 369(1-2), 869–898 (2017)
Jost, J., Xin, Y.-L.: A Bernstein theorem for special Lagrangian graphs. Calc. Var. Partial Differ. Equ. 15, 299–312 (2002)
Joyce, D.: Conjectures on Bridgeland stability for Fukaya categories of Calabi–Yau manifolds, special Lagrangians, and Lagrangian mean curvature flow. EMS Surv. Math. Sci. 2(1), 1–62 (2015)
Joyce, D.: Special Lagrangian 3-Folds and Integrable Systems. Surveys on Geometry and Integrable Systems. Advanced Studies in Pure Mathematics, vol. 51, pp. 189–233. Math. Soc. Japan, Tokyo (2008)
Joyce, D.: Lectures on Special Lagrangian Geometry. Global Theory of Minimal Surfaces. Clay Mathematics Proceedings, vol. 2, pp. 667–695. Amer. Math. Soc, Providence (2005)
Joyce, D., Lee, Y.-I., Schoen, R.: On the existence of Hamiltonian stationary Lagrangian submanifolds in symplectic manifolds. Am. J. Math. 133(4), 1067–1092 (2011)
Leung, N.C., Yau, S.-T., Zaslow, E.: From special Lagrangian to Hermitian–Yang–Mills via Fourier–Mukai transform. Adv. Theor. Math. Phys. 4(6), 1319–1341 (2000)
Nadirashvili, N., Vladut, S.: Singular solution to the special Lagrangian equations. Ann. Inst. H. Poincare Anal. Non Lineaire 27(5), 1179–1188 (2010)
Neves, A.: Singularities of Lagrangian mean curvature flow: zero-Maslov class case. Invent. Math. 168(3), 449–484 (2007)
Neves, A.: Recent Progress on Singularities of Lagrangian Mean Curvature Flow. Surveys in Geometric Analysis and Relativity. Advanced Lectures in Mathematics (ALM), vol. 20, pp. 413–438. Int. Press, Somerville (2011)
Neves, A.: Finite time singularities for Lagrangian mean curvature flow. Ann. Math. (2) 177(3), 1029–1076 (2013)
Rubinstein, Y., Solomon, J.: The degenerate special Lagrangian equation. Adv. Math. 310, 889–939 (2017)
Smith, G.: Special Lagrangian curvature. Math. Ann. 355(1), 57–95 (2013)
Smith, G.: The non-linear Dirichlet problem in Hadamard manifolds. arXiv:0908.3590
Smith, G.: The non-linear Plateau problem in non-positively curved manifolds. Trans. Am. Math. Soc. 365, 1109–1124 (2013)
Smith, G.: The Plateau problem for convex curvature functions. Ann. Inst. Fourier (to appear). arXiv:1008.3545
Schoen, R., Wolfson, J.: Minimizing area among Lagrangian surfaces: the mapping problem. J. Differ. Geom. 58, 1–86 (2001)
Schoen, R., Wolfson, J.: Minimizing Volume Among Lagrangian Submanifolds. Differential Equations: La Pietra 1996 (Florence). Proceedings of Symposia in Pure Mathematics, vol. 65, pp. 181–199. Amer. Math. Soc., Providence (1999)
Strominger, A., Yau, S.-T., Zaslow, E.: Mirror symmetry is T-duality. Nucl. Phys. B 479(1–2), 243–259 (1996)
Takahashi, R.: Tan-concavity property for Lagrangian phase operators and applications to the tangent Lagrangian phase flow. arXiv:2002.05132
Wang, M.-T.: Some Recent Developments in Lagrangian Mean Curvature Flows. Surveys in Differential Geometry, Geometric flows, vol. X II. Int. Press, Somerville (2008)
Wang, D., Yuan, Y.: Singular solutions to the special Lagrangian equations with subcritical phases and minimal surface systems. Am. J. Math. 135(5), 1157–1177 (2013)
Wang, D., Yuan, Y.: Hessian estimates for special Lagrangian equations with critical and supercritical phases in general dimensions. Am. J. Math. 136, 481–499 (2014)
Warren, M.: Calibrations associated to Monge–Ampère equations. Trans. Am. Math. Soc. 362, 3947–3962 (2010)
Wolfson, J.: Lagrangian homology classes without regular minimizers. J. Differ. Geom. 71, 307–313 (2005)
Yuan, Y.: A Bernstein problem for special Lagrangian equations. Invent. Math. 150(1), 117–125 (2002)
Yuan, Y.: Global solutions to special Lagrangian equations. Proc. A.M.S. 134(5), 1355–1358 (2005)
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by A. Neves.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Partially supported by the NSF.
Appendices
Appendix A. A Geometric interpretation of the inhomogeneous DP
The Eq. (A.1) below appeared as equation (2.18) in [21]. We left the proof as a exercise for the reader. However, this equation has an immediate consequence for the Dirichlet problem for the inhomogeneous SL equation (A.2), which is discussed in Sect. 5. This is given in Corollary A.2. It may have gone unnoticed and seems not to be well understood. For the convenience of the reader we give the proof of equation (2.18) in [21] here.
Proposition A.1
Let X be a Calabi-Yau manifold of complex dimension n. Let \(\Phi \) be the parallel (n, 0)-form normalized so that \({\mathrm{Re}} \Phi \) has comass 1. Given \(L\subset X\) an oriented Lagrangian submanifold, define the phase \(\theta \) mod \(2\pi \) by
Then for any tangent vector field V on L, we have
that is
where H is the mean-curvature vector field of L, and J is the complex structure on X.
Proposition A.1 has the following immediate implication for the inhomogeneous SL potential equation \({\mathrm{tr}}\left\{ \arctan (D^2_x u) \right\} = \psi (x)\). Let \(z\equiv x+iy \in {{\mathbb {R}}}^n\oplus i {{\mathbb {R}}}^n= {{\mathbb {C}}}^n\)
Corollary A.2
Suppose \(L\equiv \{(x, \nabla u(x)) : x\in \Omega \}\) is the graph of the gradient of \(u\in C^2(\Omega )\) over a domain \(\Omega \subset {{\mathbb {R}}}^n\). Then the inhomogeneous term
considered as a function on L, is the phase function for L. Thus it has gradient related to the mean curvature vector field H of L by
Otherwise said, if u is a solution to the equation
on \(\Omega \), with \(\psi (x)\) smooth, then
where \(\widetilde{\psi }\) is the pull-back of \(\psi \) to L.
Note. Proposition A.1 is actually independent of the orientation of L. A change of orientation changes the function \(\theta \) to \(\theta +\pi \), and the conclusion is the same. In Corollary A.2, L is given the orientation of \(\Omega \).
Proof of Proposition A.1
By a complex linear change of coordinates we may assume at p we have \(\Phi = dz_1\wedge \cdots \wedge dz_n\). Now let \(p=(x_0, \nabla u(x_0))\). The map \(D^2 u\) is symmetric, so by a change of variables \((x,y) \rightarrow (gx, gy)\) for some \(g\in {\mathrm{SO}}(n)\), we can assume that at \(x_0\), \(D^2 u\) is diagonal, i.e., \((D^2_{x_0} u)(\epsilon _k) = \lambda _k \epsilon _k\) for an orthonormal basis \(\epsilon _1,\ldots , \epsilon _n\) of \({{\mathbb {R}}}^n\).
Now let \(e_1,\ldots , e_n\) be an oriented orthonormal frame field on L in a neighborhood of \(p=(x_0, \nabla u(x_0))\). Then
and so
since \(\Phi \) is parallel. We may assume \(\nabla ^L_{e_j} e_k = (\nabla _{e_j} e_k)^{{\mathrm{tang}}} = 0\) at \(x_0\), so \(\nabla _V e_k = ( \nabla _V e_k)^{{\mathrm{normal}}} = B_{V, e_k}\) the second fundamental form of L at \(x_0\). Therefore we have
and we can write
Now pick the frame field at \(x_0\) to be
so that at \(x_0\) the vectors \(e_k\) and \(Je_k\) lie in the \(\hbox {k}\)th complex coordinate line. Recall that at the point p, \(\Phi = dz_1\wedge \cdots \wedge dz_n\). Hence, at p
Hence, with summation convention,
\(\square \)
Appendix B. The sharpness of our boundary convexity
Part of the point of this appendix is to show that strict \(\overrightarrow{\mathbf{F}}_\theta \)-convexity of the boundary \(\partial \Omega \) is the right, i.e., borderline condition for the Dirichlet problem. This goes back to work in [2] for \(\theta \) in the highest interval. We also discuss how this convexity relates to convexity of the domain \(\Omega \).
Let
-
\(\Omega \subset \subset {{\mathbb {R}}}^n\) be a domain with smooth boundary \(\partial \Omega \),
-
\(\mathbf{F}\subset {\mathrm{Sym}}^2({{\mathbb {R}}}^n)\) be a subequation,
-
\(\lambda >0\) be such that \((-{{{\mathcal {P}}}}) \cap (\lambda I+ \mathbf{F}) = \emptyset \) (this always exists), and
-
\(\varphi = -{\lambda \over 2} \Vert x\Vert ^2\bigr |_{\partial \Omega }\).
Definition B.1
\(\Omega \) is said to be \(\mathbf{F}(\Omega )\)-convex if for every \(K\subset \subset \Omega \), one has \({\widehat{K}}^{\mathbf{F}(\Omega )}\subset \subset \Omega \) where \({\widehat{K}}^{\mathbf{F}(\Omega )}{\mathop {=}\limits ^{{\mathrm{def}}}}\{x\in \Omega : f(x) \le \sup _K f, \ \forall \, f\in \mathbf{F}(\Omega )\}\)
Theorem B.2
Assume there exists a function \(h\in C^2({\overline{\Omega }})\) such that
-
(i)
\(h\bigr |_{\Omega } \in \mathbf{F}(\Omega )\),
-
(ii)
\(h\bigr |_{\partial \Omega } = \varphi \).
Then the domain \(\Omega \) is \(\mathbf{F}\)-convex.
Furthermore, let \(h^\star {\mathop {=}\limits ^{{\mathrm{def}}}}h + {\lambda \over 2} \Vert x\Vert ^2\). Then for every \(x\in \partial \Omega \) where \((Dh^\star )_x \ne 0\), there exits \(A\in ( \lambda I + \mathbf{F})\) such that
where \(c_x>0\) and \(B_x\) is the second fundamental form of \(\partial \Omega \) at x with respect to the interior normal.
In particular, if \(\mathbf{F}= \overrightarrow{{\mathbb {G}}}\) for a subequation \({{\mathbb {G}}}\), then \(\partial \Omega \) is strictly \({{\mathbb {G}}}\)-convex.
Proof
We begin by proving the second assertion. Fix \(p\in \partial \Omega \) and w.l.o.g. assume p is the origin. Choose coordinates \(x= (x', x_n)\) such that in a neighborhood of 0
where g is \(C^\infty \) with
where \(B \ {\mathop {=}\limits ^{{\mathrm{def}}}}\ \text {the second fundamental form of }\partial \Omega \text { at }0 \text {w.r.t. the interior normal}\).
In a neighborhood U of we have a defining function for \(\Omega \) given by
with
Lemma B.3
If \(\widetilde{\rho }\) is any other defining function for \(\Omega \) in U, then
Proof
In a neighborhood of 0 we have that \(\widetilde{\rho }(x) = a(x)\rho (x)\) where \(a>0\). Now
At \(x=0\) we have \(\rho (0)=0\) and \((D\rho )_0=(0,\ldots , 0, -1) \equiv n\). Therefore,
\(\square \)
Now \(\varphi = -{\lambda \over 2}\Vert x\Vert |^2\bigr |_{\partial \Omega }\), and \( h\in C^2({\overline{\Omega }})\) is \(\mathbf{F}\)-subharmonic on \(\Omega \) with boundary values \(\varphi \). That is,
-
(1)
\(D^2 h \ \in \ \mathbf{F}\) on \({\overline{\Omega }}\),
-
(2)
\(( h-\varphi )\bigr |_{\partial \Omega } = 0\).
Recall that \( h^\star \ \equiv \ h + \frac{\lambda }{2} \Vert x\Vert ^2 \) Note that \(h^\star \in C^2({\overline{\Omega }})\) is an \(\mathbf{F}^\star \)-subharmonic function for the subequation
with
Note also that \( D^2 h^\star \ \in \ \lambda I+\mathbf{F}\ \subset \ {\mathrm{Int}}\mathbf{F}\) by the positivity condition. Hence, \(h^\star \) is strictly \(\mathbf{F}\)-subharmonic.
With the supposition that \( (Dh^\star )_x \ne 0 \). we have that \(h^\star \) is a defining function for \(\partial \Omega \) in a neighborhood of x. We now apply Lemma B.3 to establish the second assertion.
Now for the first assertion. By the definition of \(\lambda \) and Theorem 3.1 in [SMP] we know that \(h^\star \) satisfies the Strong Maximum Principle, and so \(h^\star <0\) on \(\Omega \). Suppose \(K\subset \subset \Omega \) and let \(\delta \equiv {\mathrm{dist}}(K,\partial \Omega ) <0\). Then we have \({\widehat{K}}^{\mathbf{F}(\Omega )} \subset \Omega _\delta {\mathop {=}\limits ^{{\mathrm{def}}}}\{x\in \Omega : h^\star (x)\le \delta \}\). \(\square \)
Rights and permissions
About this article
Cite this article
Harvey, F.R., Lawson, H.B. Pseudoconvexity for the special Lagrangian potential equation. Calc. Var. 60, 6 (2021). https://doi.org/10.1007/s00526-020-01850-1
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00526-020-01850-1