Composite Density from the Rule of Mixtures

Of all the rules of mixtures on this site, this is the only one that is not a bound. Modulus averaging depends on an assumption about how strain or stress is shared, and the assumption is what makes the answer an upper or lower limit rather than a fact. Mass has no such difficulty. A cubic centimetre of laminate contains $V_f$ cubic centimetres of fibre and $1 - V_f$ of resin, and its mass is the sum of theirs. $\rho_c = \rho_f V_f + \rho_m (1 - V_f)$ is an identity for a void-free part, exact to as many figures as you know the inputs.

That exactness is what makes it a shop-floor tool rather than a desk one. Weigh a cured panel, measure its volume — water displacement is the honest way — and compare the measured density against this one. The shortfall is porosity, and the bookkeeping is direct: one percent by volume of voids shows up as one percent of missing density. There is no more sensitive routine measurement of laminate quality that does not involve destroying the part.

And void content matters far more than the number suggests, because voids live in the resin and the resin carries every load the fibres do not. Interlaminar shear strength falls by roughly seven percent for each percent of voids by volume; transverse strength and compressive strength follow. Above about two percent, most aerospace specifications reject the part outright. A 2% void content sounds like a rounding error and is a structural defect.

The other everyday use is weight estimation, and it is the first calculation of any composite design: a ply schedule and a surface area become a mass only once this equation has turned volume fractions into kilograms per cubic metre. It is also the denominator of both figures of merit that justify the material at all — specific stiffness and specific strength — so an error here propagates straight into the argument for using composites in the first place.

Solving backwards for $V_f$ gives a quick volume fraction from a density measurement, and it is worth being clear about its one blind spot. The inversion assumes no voids. Every bubble makes the panel lighter, and the equation reads that lightness as less fibre, so a porous laminate returns a volume fraction that is too low. That makes the density route a screening tool rather than a measurement. The standard methods do it properly — burn-off for glass, acid digestion for carbon — and both hand you the void content as a by-product, precisely by comparing the measured density against this equation.

One case where the density route fails outright is worth knowing: it only works when the two constituents differ appreciably in density. Carbon at about 1.8 g/cm³ against epoxy at about 1.2 is a comfortable gap. Aramid at about 1.44 against the same epoxy is not, and in an aramid laminate the whole span of volume fraction from zero to one moves the density across a range comparable to the measurement scatter. The equation will still return a number; it will simply not mean much.