Validity of Effective Potentials in Crowded Solutions of Linear and Ring Polymers with Reversible Bonds

Abstract

We perform simulations to compute the effective potential between the centers-of-mass of two polymers with reversible bonds. We investigate the influence of the topology on the potential by employing linear and ring backbones for the precursor (unbonded) polymer, finding that it leads to qualitatively different effective potentials. In the linear and ring cases the potentials can be described by Gaussians and generalized exponentials, respectively. The interactions are more repulsive for the ring topology, in analogy with known results in the absence of bonding. We also investigate the effect of the specific sequence of the reactive groups along the backbone (periodic or with different degrees of randomness), establishing that it has a significant impact on the effective potentials. When the reactive sites of both polymers are chemically orthogonal so that only intramolecular bonds are possible, the interactions become more repulsive the closer to periodic the sequence is. The opposite effect is found if both polymers have the same types of reactive sites and intermolecular bonds can be formed. We test the validity of the effective potentials in solution, in a broad range of concentrations from high dilution to far above the overlap concentration. For this purpose, we compare simulations of the effective fluid and test particle route calculations with simulations of the real all-monomer system. Very good agreement is found for the reversible linear polymers, indicating that unlike in their nonbonding counterparts many-body effects are minor even far above the overlap concentration. The agreement for the reversible rings is less satisfactory, and at high concentration the real system does not show the clustering behavior predicted by the effective potential. Results similar to the former ones are found for the partial self-correlations in ring/linear mixtures. Finally, we investigate the possibility of creating, at high concentrations, a gel of two interpenetrated reversible networks. For this purpose we simulate a 50/50 two-component mixture of reversible polymers with orthogonal chemistry for the reactive sites, so that intermolecular bonds are only formed between polymers of the same component. As predicted by both the theoretical phase diagram and the simulations of the effective fluid, the two networks in the all-monomer mixture do not interpenetrate, and phase separation (demixing) is observed instead.

Entropy-driven phase behavior of associative polymer networks

“…If the reactive monomers are all of the same type, in the scaling limit polymer theories predict the absence of phase separation [10]: as the polymer concentration increases, polymers start to interact with each other, and intra-molecular bonds are swapped for inter-molecular bonds, forming a (fully-bonded) transient polymer network without encountering any phase separation. Experimental [17,18] and numerical [19][20][21] evidence supports these theoretical predictions.…”

“…Since the architecture of the polymeric objects has an important role in dictating their collective behaviour, we also simulate systems made of rings decorated with N r = 24 reactive monomers arranged in an ordered manner. Recent numerical results have shown that unentangled ring polymers decorated with a single type of reactive monomers behave like chains, in that they do not experience a gas-liquid phase separation [20]. We confirm that this is the case also for our model: Figure 4(a) shows that the equation of state of the A 24 ring system in the fully-bonded limit is a monotonically-increasing function of the density, with the pressure increasing faster with density compared to an equivalent system made of chains.…”

A Reactive Bead–Spring Model for Associative Polymer Melts In and Out of Equilibrium

“…The rate of swapping is tuneable by adjusting the energy barrier for bond-swapping. This method has been utilized to study the dynamics of vitrimers − , and reversible polymer gels with different architectures . This minimal model is easily implemented and captures the essential physics of associative bonding, but its functional form is somewhat limited to forming binary complexes between two complementary species of sticky monomers and cannot be easily generalized to more complicated associative complexes like metal–ligand coordination.…”

“…Many studies have turned to molecular simulations to relate the molecular structure of networks to their viscoelastic dynamics. Simulation studies model associative bond kinetics through hybrid molecular dynamics (MD)/Monte Carlo (MC) simulation − and pure MC − or MD methods. − However, many of these models require making prescriptive assumptions about associative kinetics and how they behave in nonequilibrium flow conditions. For example, models using MC steps to perform associative reactions require user-specified reaction rates for the association and disassociation processes.…”