On the Elasticity of Polymer Model Networks Containing Finite Loops

TL;DR

This paper investigates how finite loops in polymer networks affect their elastic modulus, showing that pending loops reduce the modulus and proposing bounds based on network topology and force balance considerations.

Contribution

It introduces a new approach to estimate the phantom modulus of polymer networks with finite loops, improving upon the resistor analogy by incorporating force balance at junctions.

Findings

Pending loops reduce network modulus.

Resistor analogy provides only an approximation.

Lower bound estimate for phantom modulus derived.

Abstract

Based upon the resistor analogy and using the ideal loop gas approximation(ILGA) it is shown that only pending loops reduce the modulus of an otherwise perfect network made of monodisperse strands and junctions of identical functionality. Thus, the cycle rank of the network with pending structures removed (cyclic and branched) is sufficient to characterize modulus, if the resistor analogy can be employed. It is further shown that it is impossible to incorporate finite cycles into a polymer network such that individual network strands are at equilibrium conformations while maintaining simultaneously a force balance at the junctions. Therefore, the resistor analogy provides only an approximation for the phantom modulus of networks containing finite loops. Improved approaches to phantom modulus can be constructed from considering a force balance at the junctions, which requires knowledge of the distribution of cross-link fluctuations in imperfect networks. Assuming loops with equilibrium conformations and a force balance at all loop junctions, a lower bound estimate for the phantom modulus, $G_{\text{ph}}\approx\left(\xi-c_{\text{f}}L_{1}\right)kT/V$ is obtained within the ILGA for end-linked model networks and in the limit of $L_{1}\ll\xi$. Here, $L_{1}$ is the number of primary ("pending") loops, $\xi$ the cycle rank of the network, $k$ the Boltzmann constant, $V$ the volume of the sample, and $T$ the absolute temperature. $c_{\text{f}}$ is a functionality dependent coefficient that is $\approx2.56$ for junction functionality $f=3$ and $\approx3.06$ for $f=4$, while it converges quickly towards $\approx4.2$ in the limit of large $f$. Further corrections to phantom modulus beyond finite loops are addressed briefly.