Is it true that, in general, a good thermal conduct is also a good electrical conductor? If so, why? Does material density play some role in defining thermal/ electrical conductivity?
Edge case can be ignored, though those would be interesting to.
1. No, there is no general rule that good thermal conductor will always be a good electrical conductor. The electrical conductivity (S) is related to thermal conductivity (k) only in metals (Wiedemann-Franz law k/S = L/T) but not in insulators and semiconductors.
2. The reason of no direct relation of the 'k' with 'S' in case of non metals is that almost all amount of heat is transferred through lattice vibrations (also a function of temperature) alone in insulators and the presence of free electrons complicates the situation in semiconductors. Therefore, both contributions (phonons and free electrons that again depend upon doping and temperature) must be accounted in case of semiconductors.
3. Material density has no direct relation with 'S' of solid state materials as number of free electrons determines the 'S', not the density of material. Therefore denser material can have lower 'S'.
4. Thermal conductivity is directly related with material density.
Thermal Conductivity k = (1/3)*cs*lambda*SQRT(Y*d)
cs= specific heat, d=density, Y=Young's modulus, lambda= mean free path of phonon.
The inter-atomic distance changes with density thereby changing the velocity of the vibrations and heat capacity of the material. The velocity of thermal vibrations (speed of phonons or sound) and 'lambda' are inversely proportional to 'd' provided all other parameters are constant (especially Y). It would result in an increase in thermal conductivity of solids if the density of solid is reduced. As (mass) specific heat remains constant with density.