Abstract

Dynamic Monte Carlo simulation of a bead-spring model of flexible macromolecules threading through a very narrow pore in a very thin rigid membrane are presented, assuming at the cis side of the membrane a purely repulsive monomer-wall interaction, while the trans side is attractive. Two choices of monomer-wall attraction epsilon are considered, one choice is slightly below and the other slightly above the "mushroom to pancake" adsorption threshold epsilon(c) for an infinitely long chain. Studying chain lengths N=32, 64, 128, and 256 and varying the number of monomers N(trans) (time t=0) that have already passed the pore when the simulation started, over a wide range, we find for epsilon<epsilon(c) (nonadsorbing case) that the translocation probability varies proportional to c(trans)=N(trans)(t=0)/N for small c(trans), while for epsilon>epsilon(c) a finite number N(trans)(t=0) suffices that the translocation probability is close to unity. In the case epsilon<epsilon(c), however, the time it takes for those chains to get through the pore to complete the translocation process scales as tau proportional, variant N(2.23+/-0.04). This result agrees with the suggestion of Chuang, Kantor, and Kardar [Phys. Rev. E 65, 011802 (2001)] that the translocation time is proportional to the Rouse time, that scales under good solvent condition as tau(Rouse) proportional, variant N(2nu+1), with the excluded-volume exponent nu approximately 0.59 in d=3 dimensions. Our results hence disagree with the suggestions that the translocation time should scale as either N(2) or N(3). For epsilon>epsilon(c), we find that the translocation time scales as tau proportional, variant N(1.65+/-0.08). We suggest a tentative scaling explanation for this result. Also the distribution of translocation times is obtained and discussed.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call