Updated Effective Thickness Derivation
The paper I found postulates that the rarefaction wave will travel to the left with a velocity of $C_h$ and be replaced by an exponential decay function of $e^{-\frac{x}{C_ht}}$. Obviously, this is only valid up until $d/2 = C_ht$, at which point the wave will have converted the entire initial distribution into an exponential decay function; let us call the value for time that makes this equation true $\tau$. If we assume that the wave travels up to $x_f = 5.517C_h\tau$ (as shown in prior derivations), we can find the amount of mass that can be accounted for within the distribution: