Improved Algorithm for Identifying Partial Saddle-Points in Polymer Field Theoretic Simulation

Field-theoretic simulations that rely on a partial saddle-point approximation have become powerful tools for studying the equilibrium behavior of block copolymer materials. The computational cost of such simulations depends critically upon the efficiency of the iterative algorithm that is used to identify a partial saddle-point field configuration during each step of a stochastic simulation. We introduce a new algorithm for this purpose that relies on a physically motivated approximation in which the linear response of the density to a small change in a pressure-like field is approximated by the response of a homogeneous system. The resulting algorithm significantly reduced overall computation cost relative to the cost of the widely used Anderson-mixing algorithm.