Poisson's Ratio

Why Poisson's ratio matters

Poisson's ratio is one of exactly two independent numbers you need to fully describe how an isotropic material stretches, shears, and dilates. Young's modulus E tells you how stiff a material is along the load; ν tells you what happens sideways. That sideways coupling is not a curiosity — it drives real design decisions.

• Pressure vessels and pipes. Axial stress produces hoop strain and vice versa; ν sets the biaxial coupling that determines wall growth and end-cap deflection.
• Press fits and seals. An O-ring (ν ≈ 0.5) squeezed axially expands radially to seal a gland; a cork (ν ≈ 0) does not, which is precisely why it works as a stopper.
• Constrained geometries. A rubber pad bonded between steel plates cannot expand sideways, so its effective stiffness soars — a near-incompressible material trapped in a can behaves almost rigidly.
• Strain-gauge and FEA calibration. Every finite-element material card needs E and ν; a wrong ν skews computed stresses, especially in triaxial states.
• Auxetic engineering. Negative-ν foams and lattices resist indentation, absorb energy, and form synclastic (dome) curvature — useful in body armor, filters, and biomedical stents.

How it works, step by step

Apply a uniaxial tensile stress σ along the x-axis of a bar. The material responds with an axial strain and, in general, a lateral contraction:

• Axial strain. ε_x = σ / E, where E is Young's modulus (units Pa). This is Hooke's law along the load.
• Lateral strain. ε_y = ε_z = −ν ε_x = −ν σ / E. The minus sign says the bar contracts across the load when it extends along it.
• Poisson's ratio. ν = −ε_lateral / ε_axial. Dimensionless, because it is a ratio of two strains.
• Volume change. To first order, ΔV/V = ε_x + ε_y + ε_z = ε_x(1 − 2ν) = (1 − 2ν) σ / E. A load always adds volume unless ν = 0.5.

Notice the consequence: if ν = 0.5, the volume term vanishes and the material is incompressible — it can only change shape, not size. If ν = 0, stretching leaves the cross-section untouched and every bit of axial strain becomes volume gain. Real metals sit in between, near ν = 0.3: each side contracts by 30% of the axial strain, and the (1 − 2ν) = 0.4 that survives shows up as a net volume increase.

The elastic-constant web

For a homogeneous, isotropic, linear-elastic solid, only two of the four common constants (E, G, K, ν) are independent. Poisson's ratio is the glue:

G = E / [2(1 + ν)]   ·   K = E / [3(1 − 2ν)]   ·   E = 2G(1 + ν) = 3K(1 − 2ν)

where G is the shear modulus (Pa), K is the bulk modulus (Pa), and E is Young's modulus (Pa). Reading these off tells you the physical bounds directly: for K to be finite and positive, 1 − 2ν must be positive, so ν < 0.5. For G to be positive, 1 + ν must be positive, so ν > −1. Together they give the stability window −1 < ν < 0.5 for any isotropic material.

Poisson's ratio across materials

The number is remarkably clustered for metals and remarkably spread for everything else. Representative room-temperature values:

Material	Poisson's ratio ν	Young's modulus E (GPa)	Note
Cork	≈ 0.0	≈ 0.02	Cells buckle without spreading
Concrete	0.10 – 0.20	17 – 30	Brittle, porous
Glass (silica)	0.20 – 0.25	65 – 90	Covalent network
Cast iron	0.21 – 0.26	90 – 160	Graphite flakes lower ν
Steel	0.27 – 0.30	190 – 210	Default ν ≈ 0.3
Titanium	0.32 – 0.34	105 – 120	Aerospace alloy Ti-6Al-4V
Aluminum	0.33	68 – 72	Common structural metal
Copper	0.34	110 – 128	FCC lattice
Lead	0.44	16	Soft, near-incompressible
Gold	0.42 – 0.44	78	Very ductile FCC metal
Rubber / elastomer	0.49 – 0.4999	0.01 – 0.1	Effectively incompressible
Auxetic foam / lattice	−0.1 to −0.8	varies	Widens when stretched

Worked example: elastic constants of a steel bar

Take a structural steel with E = 200 GPa and ν = 0.30. Compute its shear and bulk moduli, and the volume change under a 250 MPa tensile stress.

• Shear modulus. G = E / [2(1 + ν)] = 200 / [2(1.30)] = 200 / 2.60 = 76.9 GPa.
• Bulk modulus. K = E / [3(1 − 2ν)] = 200 / [3(0.40)] = 200 / 1.20 = 166.7 GPa.
• Axial strain. ε_x = σ / E = 250 MPa / 200 000 MPa = 1.25 × 10⁻³ (0.125%).
• Lateral strain. ε_y = −ν ε_x = −0.30 × 1.25 × 10⁻³ = −3.75 × 10⁻⁴ (the bar narrows by 0.0375%).
• Volume change. ΔV/V = (1 − 2ν) σ/E = 0.40 × 1.25 × 10⁻³ = 5.0 × 10⁻⁴ (0.05% dilation).

Now repeat for a nearly incompressible elastomer with ν = 0.4999: the (1 − 2ν) factor collapses to 0.0002, so the same relative stress produces essentially zero volume change — all of the deformation is shape change. That single factor, (1 − 2ν), is why rubber bushings can carry enormous confined loads while quietly shearing to absorb vibration.

Common misconceptions and failure modes

• "ν must be positive." No — auxetic materials have ν < 0, valid down to −1 for isotropic solids.
• "ν = 0.5 is a normal material value." It is a limit (infinite bulk modulus). Real rubbers are 0.49–0.4999, never exactly 0.5, or the math for K blows up.
• "Volume is conserved in elasticity." Only at ν = 0.5. For steel a tensile load actually increases volume by ~(1−2ν)ε.
• "ν is the same in every direction." Only for isotropic materials. Composites, wood, and single crystals are anisotropic; their apparent ν varies with direction and can even exceed the isotropic bounds.
• "Incompressible means rigid." A confined near-incompressible rubber acts stiff, but unconfined it is soft in shear. FEA that ignores this (using standard elements for ν → 0.5) suffers volumetric locking and predicts spuriously high stiffness — use hybrid or reduced-integration elements.
• "ν has units." It is dimensionless — a pure ratio of strains.

How it is measured

The classic method mounts a biaxial strain-gauge rosette on a tensile coupon: one gauge reads ε_axial, a perpendicular gauge reads ε_lateral, and ν is the negative slope of lateral versus axial strain in the elastic region (ASTM E132). Modern labs use digital image correlation (DIC) to track a speckle pattern optically, or ultrasonic pulse-echo, where ν follows from the longitudinal and shear wave speeds: ν = (v_L² − 2v_S²) / [2(v_L² − v_S²)]. Dynamic (ultrasonic) values run slightly higher than static ones for the same material because they probe adiabatic, small-amplitude response.