Modelling the morphology and thermomechanical behaviour of low-bandgap conjugated polymers and bulk heterojunction films

1.2 Melt Phase System Initialization. Isolated chains in the fully extended conformation were first subjected to elevated temperature (800 K) simulations using Langevin dynamics (damping parameter = 3800 fs, time step = 2 fs). Random conformations were outputted on nanosecond intervals. This protocol resulted in 60 uncorrelated chain conformation for each polymer simulated. These chains were next assigned a randomly generated centre of mass and orientation with respect to an orthorhombic, periodic simulation box at a mass density of 0.01 g cm. Placement of chains was constrained such that no two atoms on different chains were within 20 Å of another. This was especially important for simulations containing PC71BM, as to avoid spearing the fullerene cage, which would result in unstable dynamics. The randomly generated configurations were next subjected to a conjugategradient energy minimization to further ensure that there were no unphysical interactions to make the simulation unstable in the first couple of time steps. Next, the temperature was ramped from 0 to 800 K over the course of 1 ns using Langevin dynamics (damping parameter = 3800 fs, dielectric constant = 9.8). Next, a Nosé-Hoover style barostat (time constant = 1000 fs) was used to relax the simulation box at 1 atmosphere of pressure until the density converged to the equilibrium melt-phase value at 800 K and 1 atm over the course of 5 ns. For all bulk heterojunction simulations, a mass fraction of 1:1.5 polymer: PCBM was used. This composition is consistent with experimentally optimized devices.