The Differential Effective Medium
Learning objectives
- Read the differential effective medium as Kuster-Toksoz's fix for crowding: add pores a pinch at a time, re-solving in the current effective medium at each step
- State the governing idea: each new sliver of porosity is embedded in the rock as it now stands, so the answer stays self-consistent to high porosity
- Confirm the endpoints and the ladder: exact mineral at zero porosity, and for dry aspect ratio 0.15 pores in calcite K about 54.0, 38.8, 20.6, and 10.9 GPa at 5, 10, 20, and 30 percent
- See the asymmetry as a deliberate modeling choice: the host mineral stays connected and the pores stay isolated, right for a rock whose mineral is load-bearing everywhere
Adding the Pores a Pinch at a Time
Kuster-Toksoz broke down because it dropped every pore into pure mineral, ignoring the ones already there. The differential effective medium, DEM, repairs exactly that. Instead of inserting all the porosity at once, it inserts an infinitesimal sliver, computes the slightly softened effective medium, then inserts the next sliver into that softened medium, and repeats until the target porosity is reached. Each pore is embedded not in bare mineral but in the rock as it currently stands, so the crowding that Kuster-Toksoz missed is built in step by step. The construction is a differential equation, roughly that the change in the modulus per unit added porosity is proportional to the contrast between the inclusion and the current medium times the shape factor, divided by the remaining mineral fraction. Integrate it from zero porosity upward and you get a frame that stays self-consistent all the way.
Exact at Zero, Sane to High Porosity
DEM inherits the good behavior and drops the bad. At zero porosity there is nothing to embed and the frame is exactly the mineral. At the very first sliver of porosity it agrees with Kuster-Toksoz, because with almost no pores present the two theories describe the same dilute situation. But as porosity grows they part company. For dry pores of aspect ratio 0.15 in calcite, DEM gives a dry bulk modulus of 54.0 GPa at 5 percent, 38.8 at 10 percent, 20.6 at 20 percent, and 10.9 at 30 percent, a smooth stiffening as porosity falls. At that 30 percent point Kuster-Toksoz gave only 6.75 GPa; DEM gives 10.87, and the difference is the crowding that DEM keeps track of and Kuster-Toksoz throws away.
A Choice About Texture
DEM contains a quiet decision worth naming. It embeds pores into mineral, never mineral into pores, so the roles are asymmetric: the mineral is always the connected host and the pores are always the isolated guests. That is not a mathematical accident but a statement about the rock. It says the solid frame is continuous and load-bearing everywhere, with pores as disconnected inclusions floating inside it, which is a faithful picture of a well-cemented sandstone or a tight carbonate whose mineral never loses its grip. It would be the wrong picture for a rock near falling apart, where the pores connect and the solid stops carrying load. Holding the host connected at every porosity is the DEM signature, and it is what makes the model behave sanely even at high porosity. The next section drops that privilege entirely, letting every phase be an inclusion in the mixture, and discovers a dramatic new behavior that DEM by construction can never show.
References
- Norris, A. N. (1985). A differential scheme for the effective moduli of composites. Mechanics of Materials, 4(1), 1-16.
- Mavko, G., Mukerji, T., & Dvorkin, J. (2009). The Rock Physics Handbook (2nd ed.). Cambridge University Press.