← Back to arXiv
arXivAnalysis / PDEsarXiv:2607.12333

On the p-torsional rigidity of compact metric graphs: a sharp Kohler--Jobin inequality

The paper studies a mathematical quantity called "p-torsional rigidity" on compact metric graphs. A metric graph is a network of edges (like line segments) connected at vertices, where the edges have actual lengths. Torsional rigidity originally comes from physics, where it measures how resistant a rod or beam is to twisting. In mathematics, this concept has been generalized to abstract settings, and here it describes a property of how solutions to a certain nonlinear equation (involving the p-Laplacian operator) behave on these graph structures. The boundary conditions imposed mean that the function is forced to be zero at certain endpoints of the graph, while interior junction points satisfy a kind of flow-balance condition.

The authors prove two main geometric inequalities. The first is a p-Saint-Venant inequality, which says that among all compact metric graphs with a fixed total length, the torsional rigidity is largest when the graph is just a single interval with one endpoint fixed to zero. In other words, a simple line segment is the most "rigid" shape for a given length. The second result is a p-Kohler-Jobin inequality, which connects torsional rigidity to the first eigenvalue of the p-Laplacian (a number that captures the fundamental vibrational frequency of the graph in a generalized sense). Specifically, it gives a lower bound on this eigenvalue that depends on the torsional rigidity in a way that is consistent across different scales.

Both inequalities are sharp, meaning there exist specific graphs that achieve equality, so the bounds cannot be improved. These results generalize classical inequalities from geometry and mathematical physics, originally known for flat two-dimensional domains, to the setting of networks. They also extend earlier work on metric graphs that only handled the linear case (p equals 2) to all values of p greater than 1. The paper contributes to a broader program of understanding how the shape and connectivity of a network influence the behavior of nonlinear differential equations defined on it.

Read original →