Surface Segregation in Athermal Polymer Blends Due to Conformational Asymmetry

Abstract

Monte Carlo simulations are used to investigate the surfaces of athermal blends of stiff and flexible polymers as the bending modulus of the stiff polymers, κ, is increased from zero to the point where the bulk undergoes an isotropic−nematic transition. For hard walls characteristic of polymer/solid surfaces, the flexible polymers generally segregate to the surface. However, just prior to the bulk transition, a surface transition occurs where a thin nematic layer rich in stiff polymers forms next to the wall. This appears to be followed by a second surface transition where the degree of order abruptly increases, which, in turn, causes a rapid thickening of the nematic layer that nucleates the bulk nematic phase. For soft boundaries representative of polymer/air surfaces, a thin layer rich in stiff polymers but without nematic order forms on the outer edge of the surface with a more significant layer rich in flexible polymers beneath. In this case, the surface transitions are lost and the qualitative behavior remains unchanged as κ increases.

Universality of Entropic Surface Segregation from Athermal Polymer Blends Due to Conformational Asymmetry

“…On the other hand, molecular dynamics (MD) simulations on diblock copolymer melts by Nikoubashman et al 19 reported a preference for the component with the shorter statistical segment length. Most recently, we performed MC simulations of flexible and stiff polymers 20 polymer, quantitatively consistent with the SCFT calculations of Wu et al 15 Despite their diametrically opposing conclusions, the only meaningful difference between the simulations of Kumar and coworkers 16,17 and our previous ones 20 is that they are based on off-lattice and lattice models, respectively. It is a common practice to restrict the monomers of large and relatively flexible polymers to a lattice so as to increase the computational efficiency of simulations, with the expectation that the artificial lattice will not alter the qualitative behavior.…”

“…The differences are most severe for the larger polymers, but they still remain significant even for their shortest N = 10 polymer. The inconsistency is readily explained by a lack of equilibration as initially suspected 20 as well as a lack of statistics. We performed many orders of magnitude more equilibration and several orders of magnitude more statistics.…”

“…Indeed, this was illustrated in our previous study of the lattice model, where simulations were performed with difference surface fields. 20 Now that we have shown F ρ (Z) to be a universal profile, the next issue will be to understand how the inner solution and thus A ρ depend on the system details, including packing effects. However, this is beyond the scope of the present paper.…”

“…Because of a general tendency for flexible polymers to segregate to the surface, a depletion occurs in the bulk region. In our previous lattice simulations, 20 we tweaked the numbers of flexible and stiff polymers to restore the 50−50 composition in the bulk region, but this is challenging for the smaller system sizes typical of offlattice simulations. For the current system size, the number of flexible polymers needs to be increased to 24.33, but of course n f must be an integer.…”

“…This high compressibility implies that the monomer density chosen by Kumar et al, nN/V = 0.6, is still significantly less than that of a real melt. 20 There are also visible oscillations showing evidence for second and third layers, but in this case they are dominated by monomers from the flexible polymers.…”