The bulk density of a material (kg/m^3) has no impact on the (mass) specific heat (J/kg/K)
Consider, for example foamed aluminium. It has a low bulk density, but its specific heat is still that of solid aluminium. The (mass) specific heat takes no account of the void fraction - it's the amount of heat per unit mass per degree that a material can hold.
Let us consider the case of a binary composite. The following 'mixture rule' can allow us to estimate the specific heat capacity of the composite, Cp. having SI units J/(kg·K)), based at additive contributions of both the filler (CpF) and the matrix (CpM): Cp ≈ (x·ρF·CpF + (1-x)·ρM·CpM) / ρ. Here, x denotes the volumetric fraction of the filler; ρ, ρF, and ρM, stand for the densities of composite, filler, and matrix, respectively. Although smaller density tends to translate itself in higher heat capacity for the composite, we must remind that both the quantities are interrelated because composition dependent. The density of the composite can also be estimated by a mixture rule, after the densities of filler and matrix: ρ ≈ x·ρF + (1-x)·ρM. Thus we may also write Cp ≈ (x·ρF·CpF + (1-x)·ρM·CpM) / [x·ρF + (1-x)·ρM], showing how the specific heat capacity of the composite can be expected to depend on the volumetric fraction of the filler.
Specific heat capacity of any material = its thermal effusivity /(density*square root of (its thermal diffusivity)). That means specific heat capacity and density are always inversely proportional. This is why, as density increases, the specific heat capacity decreases and vice versa.