A unified model for linear responses of physical networks

José M. Ortiz-Tavárez, William Stephenson, Xiaoming Mao · 2025

Read A unified model for linear responses of physical networks by José M. Ortiz-Tavárez, William Stephenson, Xiaoming Mao (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-2508-04616.txt.

Introduction

Jose Ortiz-Tavarez, William Stephenson, and Xiaoming Mao unified the physics of networks in a 2025 paper. Springs and lattices, circuits, tissues, and built metamaterials all move quantities along links. Each field grew its own math for how a network answers a push, though the skeletons match.

Their frame uses algebraic graph theory to cover them all at once. Nodes hold potentials, edges carry flows, and one set of rules yields still and moving answers alike. Heat, charge, force, and diffusion all fit the scheme. One language for many networks could speed design of new materials.

Catalog record

CREATORJosé M. Ortiz-Tavárez, William Stephenson, Xiaoming Mao
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE108 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-04616

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A unified model for linear responses of physical networks José M. Ortiz-Tavárez, William Stephenson, Xiaoming Mao Abstract Many physical systems—from mechanical lattices and electrical circuits to biological tissues and architected metamaterials—can be understood as networks transmitting physical quantities . We present a unified mathematical framework for describing linear responses of such physical networks using tools from algebraic graph theory. This approach captures static and dynamic behaviors across multiple domains, including mechanical, electrical, thermal, and diffusive responses using node and edge variables (e.g., potentials, flows). Our formalism connects multiscale and multi-domain responses to the underlying network structure. We demonstrate how this framework enables efficient, generalizable solutions to a wide class of linear response problems, including stress propagation, charge transport, and wave dynamics, and provide insights into network duality and entropy production. I Introduction All materials are fundamentally discrete at the atomic scale. As one zooms out to larger scales, microscopic heterogeneities often average out, and a homogenized continuum theory emerges as a good description of physical properties Cioranescu and Donato 1999 ; Torquato et al. 2002 ; Milton and Sawicki 2003 . For instance, in crystals, long-range order based on repeating unit cells enables elegant mathematical formulations, such as band theory. In glasses, despite the lack of long-range order, statistical rotational and translational invariance beyond a few atomic lengths allows for the construction of effective continuum theories grounded in symmetry principles. However, many materials exhibit structural complexity across multiple length scales and cannot be captured by traditional homogenization alone Lakes 1993 ; Park and Lakes 2007 ; Mao and Kotov 2024 . A prime example is biological tissue, which incorporates hierarchical organization from macromolecules and organelles to cells and vascular networks, spanning nanometers to millimeters Park and Lakes 2007 . Such multiscale materials are also ubiquitous in engineering—fiber-reinforced composites, architected metamaterials, and hierarchical porous media—where structural features span from nanometers to centimeters to achieve tailored mechanical, thermal, or acoustic properties Lakes 1993 ; Park and Lakes 2007 ; Mao and Kotov 2024 . Modeling such multiscale architectures remains a grand challenge, as conventional multiscale modeling methods are often tailored to specific materials and difficult to generalize across different material classes Lukkassen and Milton 2000 ; Bouvard et al. 2009 ; Weinan 2011 ; Ramírez-Torres et al. 2019 . In parallel, advances in network science offer a powerful alternative perspective Barabási 2013 ; Newman 2018 . Can we characterize the architecture of multiscale materials statistically—using concepts such as connectivity, community structure, geodesic distance, centrality, and spectral properties—to build efficient and generalizable theoretical models? Such an approach holds promise for capturing the essential features of complex materials where structural details vary widely but exhibit statistical regularities. Recent developments in this direction have yielded exciting new insights into systems ranging from granular matter and gels to geological materials Bassett et al. 2012 ; Zhang et al. 2017 ; Papadopoulos et al. 2018 ; Kollmer and Daniels 2019 ; Dehmamy et al. 2018 ; Zhang et al. 2019 ; Rocklin et al. 2021 ; Machlus et al. 2021 ; Vecchio et al. 2021 ; Yang et al. 2024 ; Pete et al. 2024 . In this paper, we provide a mathematical framework to describe linear responses of physical networks, which can serve as a unifying tool to accelerate discoveries in this new front. Discrete network-based models arise in many contexts in physics and engineering, from Maxwell’s mechanical frames and phonons in atomic lattices Lubensky et al. 2015 ; Mao and Lubensky 2018 , to electric circuits W.A. 1968 ; Bollobás 1998 and finite element modeling of multiphysics models Szabó and Babuška 2021 ; Zimmerman 2006 , offering a rich foundation of knowledge. Building on this, we extract universal elements from these approaches and propose a streamlined mathematical framework grounded in algebraic graph theory Swamy and Thulasiraman 1981 ; Biggs 1993 , capable of representing diverse physical phenomena, static and dynamic, in a unified language. As summarized in Table 1 , our approach maps linear responses in various domains—mechanical, electrical, thermal, diffusive—to potentials and flows on nodes and edges of a network. This mapping, combined with the linear algebraic structure of graphs that relates nodes and edges, enables an efficient and extensible framework for analyzing physical responses, coupling across domains, and connecting physical behavior to underlying network architecture. Properties Mechanical (static) Mechanical (mobility) Electrical Thermal Diffusive Potential Position \mathbf{U}_{i} Velocity \mathbf{V}_{i} Electric potential V_{j} Temperature Chemical potential Node incoming current Force \mathbf{F}_{i} Force \mathbf{F}_{i} Incoming current I_{j} Incoming heat Incoming matter Potential difference Extension e_{nm} Extension rate \dot{e}_{nm} Voltage v_{nm} Temperature difference Chemical potential difference Flow Tension t_{nm} Tension t_{nm} Current i_{nm} Heat current Matter current Transported quantity Momentum Momentum Charge Heat Matter Conductivity Stiffness Mechanical impedance Electrical admittance Thermal conductivity Diffusivity Table 1: Analogous potentials and flows on physical networks across different domains. This paper is organized as follows. In Sec. II we introduce the basic formulation of how linear responses in multiple domains (Table 1 ) can be written in a unified language based on potential-flow problem on networks. In Sec. III we discuss static responses of mechanical networks to showcase the utility of this method in efficiently calculating a diverse set of problems. In Sec. IV we discuss a set of irreversible transport problems on networks, the formulation of which runs in parallel with static problems, albeit with one time derivative. In Sec. V we discuss dynamics and wave propagation on networks. II Physical networks II.1 Notation First we define some conventions of notation that we will use in this paper. Because both network indices and Cartesian indices are necessary for mechanical networks, we adopt a unified convention of notation here. Upper indices identify columns while lower indices identify rows. Repeated indices are summed over only when they appear as both lower and upper indices. Lower case Latin indices label nodes. They can also label cycles or faces. Greek indices label spatial directions x,y,z . Upper case Latin indices label irreversible processes. These are only used in the Sec. IV . We often group indices. For example e_{ij} represents a quantity on edge ij . Here ij plays the role of an edge index and is understood to run only over edges that actually exist in the network. We adopt this notation of using the indices on the two nodes an edge connects to label the edge so it naturally denotes a directionality from i to j , which is useful for the discussion of physical transport. Consider U_{n}^{\alpha} in a two dimensional network of three nodes. The lower index n labels a row while the upper index \alpha labels a column. Written explicitly as a matrix we have: U_{n}^{\alpha}\equiv\begin{pmatrix}U_{1}^{x}&U_{1}^{y}\\ U_{2}^{x}&U_{2}^{y}\\ U_{3}^{x}&U_{3}^{y}\\ \end{pmatrix}. (1) If we instead write U_{n\alpha} , then n\alpha labels a row: U_{n\alpha}\equiv\begin{pmatrix}U_{1x}\\ U_{1y}\\ U_{2x}\\ U_{2y}\\ U_{3x}\\ U_{3y}\\ \end{pmatrix}. (2) II.2 Flows and potentials on networks We use the term physical network to refer to structures where there occurs a transmission or flow of some physical quantity through the edges, and said flow can lead to accumulation on the nodes. Examples of this transmitted quantity or flow include electric current, thermal or diffusive flow, and mechanical tension, among others. This is in contrast to non-physical networks, such as social networks,

Read the complete text