Polymer translocation through a nanopore: the effect of solvent conditions

We study the translocation of a polymer through a nanopore by means of dissipative particle dynamics (DPD). Unlike Langevin approaches, DPD explicitly takes into account the interactions of solvent and polymer. We find that the translocation time for unforced translocation follows a scaling tau approximately N(beta) with beta approximately 2.24 in good agreement with the prediction beta = 1 + 2nu that has been derived by considering hydrodynamics and memory effects within the chain. For bad-solvent conditions beta --> 2, i.e., a diffusive scaling arises as a consequence of the reduced polymer relaxation time. Biased translocation between a good and a bad-solvent reservoir (tuned via the repulsion between solvent and polymer) yields a preferential translocation toward the good solvent with beta approximately 1.2. This observation is consistent with the recent theoretical prediction beta = 3nu/(1 + nu) for driven translocation. When varying the solvent quality by imposing attractive monomer-monomer interactions (such as in Langevin approaches), an artificial translocation toward the bad-solvent side emerges. Using attractive monomer-monomer interactions to mimic a bad solvent hence does not capture the essential physics of the translocation process.