Minimal surface
In mathematics, a minimal surface is a surface that locally minimizes its area. This is equivalent to having zero mean curvature.
The term "minimal surface" is used because these surfaces originally arose as surfaces that minimized total surface area subject to some constraint. Physical models of area-minimizing minimal surfaces can be made by dipping a wire frame into a soap solution, forming a soap film, which is a minimal surface whose boundary is the wire frame. However, the term is used for more general surfaces that may self-intersect or do not have constraints. For a given constraint there may also exist several minimal surfaces with different areas : the standard definitions only relate to a local optimum, not a global optimum.
Definitions
Minimal surfaces can be defined in several equivalent ways in R3. The fact that they are equivalent serves to demonstrate how minimal surface theory lies at the crossroads of several mathematical disciplines, especially differential geometry, calculus of variations, potential theory, complex analysis and mathematical physics.This property is local: there might exist other surfaces that minimize area better with the same global boundary.
This definition makes minimal surfaces a 2-dimensional analogue to geodesics.
By the Young–Laplace equation the curvature of a soap film is proportional to the difference in pressure between the sides: if it is zero, the membrane has zero mean curvature. Spherical bubbles are not minimal surfaces as per this definition: while they minimize total area subject to a constraint on internal volume, they have a positive pressure.
A direct implication of this definition is that every point on the surface is a saddle point with equal and opposite principal curvatures.
The partial differential equation in this definition was originally found in 1762 by Lagrange, and Jean Baptiste Meusnier discovered in 1776 that it implied a vanishing mean curvature.
This definition ties minimal surfaces to harmonic functions and potential theory.
A direct implication of this definition and the maximum principle for harmonic functions is that there are no compact complete minimal surfaces in R3.
This definition uses that the mean curvature is half of the trace of the shape operator, which is linked to the derivatives of the Gauss map. If the projected Gauss map obeys the Cauchy–Riemann equations then either the trace vanishes or every point of M is umbilic, in which case it is a piece of a sphere.
The local least area and variational definitions allow extending minimal surfaces to other Riemannian manifolds than R3.
History
Minimal surface theory originates with Lagrange who in 1762 considered the variational problem of finding the surface z = z of least area stretched across a given closed contour. He derived the Euler–Lagrange equation for the solutionHe did not succeed in finding any solution beyond the plane. In 1776 Jean Baptiste Marie Meusnier discovered that the helicoid and catenoid satisfy the equation and that the differential expression corresponds to twice the mean curvature of the surface, concluding that surfaces with zero mean curvature are area-minimizing.
By expanding Lagrange's equation to
Gaspard Monge and Legendre in 1795 derived representation formulas for the solution surfaces. While these were successfully used by Heinrich Scherk in 1830 to derive his surfaces, they were generally regarded as practically unusable. Catalan proved in 1842/43 that the helicoid is the only ruled minimal surface.
Progress had been fairly slow until the middle of the century when the Björling problem was solved using complex methods. The "first golden age" of minimal surfaces began. Schwarz found the solution of the Plateau problem for a regular quadrilateral in 1865 and for a general quadrilateral in 1867 using complex methods. Weierstrass and Enneper developed more useful representation formulas, firmly linking minimal surfaces to complex analysis and harmonic functions. Other important contributions came from Beltrami, Bonnet, Darboux, Lie, Riemann, Serret and Weingarten.
Between 1925 and 1950 minimal surface theory revived, now mainly aimed at nonparametric minimal surfaces. The complete solution of the Plateau problem by Jesse Douglas and Tibor Radó was a major milestone. Bernstein's problem and Robert Osserman's work on complete minimal surfaces of finite total curvature were also important.
Another revival began in the 1980s. One cause was the discovery in 1982 by Celso Costa of a surface that disproved the conjecture that the plane, the catenoid, and the helicoid are the only complete embedded minimal surfaces in R3 of finite topological type. This not only stimulated new work on using the old parametric methods, but also demonstrated the importance of computer graphics to visualise the studied surfaces and numerical methods to solve the "period problem". Another cause was the verification by H. Karcher that the triply periodic minimal surfaces originally described empirically by Alan Schoen in 1970 actually exist. This has led to a rich menagerie of surface families and methods of deriving new surfaces from old, for example by adding handles or distorting them.
Currently the theory of minimal surfaces has diversified to minimal submanifolds in other ambient geometries, becoming relevant to mathematical physics and three-manifold geometry.
Examples
Classical examples of minimal surfaces include:- the plane, which is a trivial case
- catenoids: minimal surfaces made by rotating a catenary once around its directrix
- helicoids: A surface swept out by a line rotating with uniform velocity around an axis perpendicular to the line and simultaneously moving along the axis with uniform velocity
- Schwarz minimal surfaces: triply periodic surfaces that fill R3
- Riemann's minimal surface: A posthumously described periodic surface
- the Enneper surface
- the Henneberg surface: the first non-orientable minimal surface
- Bour's minimal surface
- the Gyroid: One of Schoen's 1970 surfaces, a triply periodic surface of particular interest for liquid crystal structure
- the Saddle tower family: generalisations of Scherk's second surface
- Costa's minimal surface: Famous conjecture disproof. Described in 1982 by Celso Costa and later visualized by Jim Hoffman. Jim Hoffman, David Hoffman and William Meeks III then extended the definition to produce a family of surfaces with different rotational symmetries.
- the Chen–Gackstatter surface family, adding handles to the Enneper surface.
Generalisations and links to other fields
The definition of minimal surfaces can be generalized/extended to cover constant-mean-curvature surfaces: surfaces with a constant mean curvature, which need not equal zero.
In discrete differential geometry discrete minimal surfaces are studied: simplicial complexes of triangles that minimize their area under small perturbations of their vertex positions. Such discretizations are often used to approximate minimal surfaces numerically, even if no closed form expressions are known.
Brownian motion on a minimal surface leads to probabilistic proofs of several theorems on minimal surfaces.
Minimal surfaces have become an area of intense scientific study, especially in the areas of molecular engineering and materials science, due to their anticipated applications in self-assembly of complex materials. The endoplasmic reticulum, an important structure in cell biology, is proposed to be under evolutionary pressure to conform to a nontrivial minimal surface.
Minimal surfaces play a role in general relativity. The apparent horizon is a minimal hypersurface, linking the theory of black holes to minimal surfaces and the Plateau problem.
Minimal surfaces are part of the generative design toolbox used by modern designers. In architecture there has been much interest in tensile structures, which are closely related to minimal surfaces. A famous example is the Olympiapark in Münich by Frei Otto, inspired by soap surfaces.
In the art world, minimal surfaces have been extensively explored in the sculpture of Robert Engman, Robert Longhurst, and Charles O. Perry, among others.