Neither a solid nor a liquid
An ideal elastic solid (a spring) stores energy and springs back instantly; an ideal viscous liquid (a dashpot) simply resists the rate of deformation and never springs back at all. Almost every real material — rubber, dough, biological tissue, molten polymer, silly putty — sits between the two, storing some energy elastically and dissipating some viscously depending on how fast you deform it. Viscoelasticity is the framework for describing that in-between behaviour, and the standard trick is to build it out of springs and dashpots combined in different arrangements.
Maxwell, Kelvin-Voigt, and the Standard Linear Solid
The Maxwell model puts a spring and a dashpot in series: under a constant strain the spring's stress relaxes away exponentially as the dashpot slowly creeps to relieve it, so it behaves like a solid on short timescales and a liquid on long ones — good for describing stress relaxation but it never fully recovers its original shape after flowing. The Kelvin–Voigt model puts them in parallel instead: applying a constant stress makes the material creep toward an equilibrium strain and then hold it, capturing delayed elastic recovery but not stress relaxation. The Standard Linear Solid (a Maxwell arm and a lone spring in parallel) combines both behaviours and is the simplest model that gets stress relaxation, creep, and full recovery all qualitatively right.
Maxwell (series): stress relaxation, σ(t) = σ₀·e^(−t/τ), τ = η/G₀ Kelvin-Voigt (parallel): creep toward equilibrium strain, no relaxation Standard Linear Solid: Maxwell arm ∥ spring — relaxation AND recovery τᴿ = relaxation time (η/G₀) sets the boundary between "acts solid" (t ≪ τᴿ) and "acts liquid" (t ≫ τᴿ)
Two experiments that separate the models
Stress relaxation: hold the strain fixed and watch the stress. A pure elastic spring holds it forever; Maxwell and the SLS let it decay toward a plateau (zero for Maxwell, a finite value for the SLS) with a characteristic time constant τᴿ = η/G₀ — the ratio of viscosity to stiffness. Creep: hold the stress fixed and watch the strain grow. Kelvin-Voigt and the SLS approach a finite equilibrium strain asymptotically; pure Maxwell creeps forever at a constant rate because the dashpot never stops extending under constant load.
Oscillating instead: the complex modulus
Apply a small sinusoidal strain instead of a step, and the resulting stress oscillates at the same frequency but shifted in phase — in phase with strain for a pure elastic solid, 90° out of phase for a pure viscous liquid, somewhere in between for a viscoelastic material. Splitting the response into its storage modulus G′(ω) (the in-phase, elastic part) and loss modulus G″(ω) (the out-of-phase, viscous part) shows the same solid/liquid crossover as a function of frequency instead of time: at low frequency a Maxwell-type material behaves like a liquid (G″ dominates), and at high frequency it behaves like a solid (G′ dominates), crossing over near ω ≈ 1/τᴿ.
Why this matters beyond the lab
Rheology built on exactly this spring-dashpot bookkeeping governs how bread dough is kneaded and proofed, how asphalt resists rutting under a hot afternoon of traffic but stays rigid overnight, how biological tissue like cartilage cushions impact, and how 3D-printing polymers need to be fluid enough to extrude yet solid enough to hold their shape the instant they are deposited — all of it explained by where a material sits on the same solid-to-liquid spectrum this simulation lets you drag through directly.
Frequently asked questions
What is the practical difference between the Maxwell and Kelvin-Voigt models?
Maxwell (spring and dashpot in series) captures stress relaxation but creeps forever under constant load and never fully snaps back. Kelvin-Voigt (spring and dashpot in parallel) captures delayed elastic recovery and creep to equilibrium but cannot show stress relaxation at all — each model is missing one of the two classic viscoelastic behaviours.
What does the relaxation time τᴿ actually represent?
It is the ratio of viscosity to stiffness, η/G₀, and it marks the timescale where the material's behaviour crosses over. Deform it much faster than τᴿ and it responds like an elastic solid; deform it much slower and it responds like a viscous liquid.
Why do storage and loss modulus depend on frequency?
Because how much of the imposed deformation gets stored elastically versus dissipated viscously depends on how fast you oscillate relative to the material's own relaxation time. Low frequency gives the dashpot time to flow (loss dominates); high frequency does not (storage dominates), so G′(ω) and G″(ω) trace out the same solid-liquid crossover as a function of frequency instead of time.
Try it live
Everything above runs in your browser — open Viscoelastic Fluid and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.
▶ Open Viscoelastic Fluid simulation