Viscoelastic Waves: When Diffusion Turns Into Propagation

Neither a pure solid nor a pure liquid

An ideal elastic solid stores mechanical energy and returns it - hit a bell and it rings. An ideal viscous liquid dissipates mechanical energy as heat - stir honey and the motion just stops. Most real materials, from biological tissue to polymers to the Earth's mantle over geological timescales, do both at once: they are viscoelastic. Which behaviour dominates in a given situation is not a fixed material property but depends on how fast you probe the material, and that time-dependence is exactly what makes viscoelastic wave propagation a richer, stranger subject than ordinary elastic wave propagation.

The Maxwell model: a spring and a dashpot in series

The simplest mathematical description that captures both instantaneous elastic response and long-term viscous flow is the Maxwell model: a spring of stiffness G in series with a viscous dashpot of viscosity eta. Stress applied suddenly is initially carried entirely by the spring (elastic response), but under sustained stress the dashpot slowly extends and the stress relaxes toward zero (viscous flow) - a real solid held under constant strain that gradually loses its resistance, which is exactly what stress relaxation in polymers and rocks looks like.

Maxwell model:  spring (modulus G) + dashpot (viscosity eta), in SERIES
relaxation time:      tau = eta / G
stress under constant strain:   sigma(t) = sigma_0 * exp(-t / tau)     // exponential stress relaxation
complex modulus:      G*(omega) = G * (i*omega*tau) / (1 + i*omega*tau)

The crossover frequency: where diffusion turns into propagation

The single most important quantity in the model is the relaxation time, tau = eta / G, or equivalently its inverse, the crossover (angular) frequency omega_c = G / eta. It marks the boundary between two qualitatively different kinds of wave behaviour in the same material:

omega << omega_c  (slow forcing, long timescale relative to tau):
    the dashpot has time to relax -> material behaves VISCOUS -> disturbances DIFFUSE, heavily damped, no clean wavefront

omega >> omega_c  (fast forcing, short timescale relative to tau):
    the dashpot has no time to move -> material behaves ELASTIC -> disturbances PROPAGATE as a wave with much less attenuation

This is why the same block of material can seem solid or liquid depending purely on how fast you hit it - a well-known party trick with silicone-based "bouncing putty" that bounces like a rigid rubber ball when thrown (fast forcing, above omega_c) but flows and puddles under its own weight if left standing (slow forcing, below omega_c). The complex modulus G*(omega) formalizes this: its real part (the storage modulus) tracks elastic energy storage and dominates at high frequency, while its imaginary part (the loss modulus) tracks viscous dissipation and dominates at low frequency, and the two cross over right at omega = omega_c.

Beyond the simplest model

The bare Maxwell model has a well-known flaw: it predicts a material flows without limit under any sustained constant stress, which is unrealistic for a genuine solid. The Kelvin-Voigt model (spring and dashpot in parallel instead of series) fixes that at the cost of predicting no instantaneous elastic response, and the standard linear solid (Zener model) combines elements of both to capture materials that both have a finite equilibrium stiffness and relax over time - a more realistic description of most soft biological tissues and polymers, at the cost of a second relaxation time to fit.

Real-world dispersion and attenuation

Viscoelastic wave models with a nonzero relaxation time necessarily predict dispersion - different frequency components of a wave travel at different speeds - and frequency-dependent attenuation, since energy loss through the dashpot scales with how fast the material is being deformed. This is directly measurable and directly useful: elastography, a medical imaging technique, sends low-frequency shear waves through tissue and measures their speed and attenuation to infer stiffness differences that can flag tumours (which tend to be stiffer than surrounding healthy tissue) noninvasively. Seismologists use essentially the same physics in reverse, at a vastly different scale: the Earth's mantle behaves elastically for fast seismic waves but flows viscously over the thousands-of-years timescale of postglacial rebound, and matching both regimes with a single viscoelastic rheology is central to modern models of mantle convection and plate tectonics.

Why the crossover matters for wave simulations

Simulating viscoelastic wave propagation numerically means picking a time step and spatial resolution fine enough to resolve the fastest frequency components you care about relative to omega_c, and getting the relaxation-time physics into the discretized equations correctly - a naive elastic wave solver with damping bolted on will get the attenuation magnitude wrong unless it actually implements the frequency-dependent complex modulus rather than a single fixed damping coefficient.

Frequently asked questions

Try it live

Everything above runs in your browser — open Viscoelastic Waves and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.