Detachment of semiflexible polymer chains from a substrate - a Molecular Dynamics investigation

TL;DR

This study uses Molecular Dynamics simulations to analyze the force profile during polymer detachment from a substrate, revealing the influence of chain stiffness, hydrodynamics, and binding energies, with validation against atomistic models.

Contribution

It introduces a comprehensive MD simulation approach to study polymer detachment, highlighting the effects of bending stiffness, hydrodynamics, and segment binding energies, and validates findings with atomistic simulations.

Findings

Oscillations depend on bending stiffness, not torsional stiffness.

Hydrodynamic interactions have minimal effect on force profiles.

Binding energy ratios significantly alter detachment force profiles.

Abstract

Using Molecular Dynamics simulations, we study the force-induced detachment of a coarse-grained model polymer chain from an adhesive substrate. One of the chain ends is thereby pulled at constant speed off the attractive substrate and the resulting saw-tooth profile of the measured mean force $< f >$ vs height $D$ of the end-segment over the plane is analyzed for a broad variety of parameters. It is shown that the observed characteristic oscillations in the $< f >$-$D$ profile depend on the bending and not on the torsional stiffness of the detached chains. Allowing for the presence of hydrodynamic interactions (HI) in a setup with explicit solvent and DPD-thermostat, rather than the case of Langevin thermostat, one finds that HI have little effect on the $< f >$-$D$ profile. Also the change of substrate affinity with respect to the solvent from solvophilic to solvophobic is found to play negligible role in the desorption process. In contrast, a changing ratio $\epsilon_s^A / \epsilon_s^B$ of the binding energies of $A$- and $B$-segments in the detachment of an $AB$-copolymer from adhesive surface strongly changes the $< f >$-$D$ profile whereby the $B$-spikes vanish when $\epsilon_s^A / \epsilon_s^B < 0.15$. Eventually, performing an atomistic simulation of a (bio)-polymer {\it polyglycine}, we demonstrate that the simulation results, derived from our coarse-grained model, comply favorably with those from the all-atom simulation.