Switch to coarsening from the Gibbs-Thompson Effect

[WardLT]

You can derive the growth rate of a particle as a function of radius as a second phase of particles grows in a super-saturated matrix. Some of the assumptions (e.g., a fixed composition) may not be true for radiation, but the form has better limiting behavior than our current model:

$\frac{d\rho}{dt} = \frac{1}{\rho}(\Theta_m - \frac{1}{\rho})$

See "Phase Transformations" by Hoyt pg 153

[WardLT]

While at it, can we verify that no voids beyond $1/\Theta_m$ shrink?