An MD study of the polymer–polymer adhesion via connector chains: some aspects of the competition between bulk dissipation in the interphase and pull-out of interface-connecting molecules
Graphical abstract
Introduction
This contribution is interested in fracture toughness and the fracture mechanisms of planar polymer interfaces. It represents an intermediate continuation of the already published work by Solar and van der Giessen [1], where a first insight of the competition between surface and bulk phenomena, involved in polymer interface separation problems, was presented for various separation velocities. Various stress measures were discussed and assessed at various locations of the heterogeneous system, as well.
Damaging or fracture in polymers is still a broad research topic. In comparison to metallic materials where fracture only consists of the propagation of a crack, the region at the front of a crack tip in a polymeric material, is made of two lips joined (on a very small characteristic distance) by isolated fibrils. This region is the so-called “craze” and is the prelude of a pure crack. This region propagates, strictly speaking, through a mechanism of “crazing” instead of “cracking”. Depending on the mode I, II or III of craze-opening, this mechanism of “crazing” is composed of the 3-D oriented translation of the front of the craze, in which one observes fibrils extraction from the lips or fibrils breaking and, after which the craze degenerates into a pure crack like in metallic materials. If one considers a schematic representation of the “crazing” propagation mechanism where a fibril is seen like a coarse-grained polymer molecule, one may find interesting similitude with the issue of polymeric interface separation, where some polymer molecules diffused at both sides of the interface.
The topic of fracture of planar polymer interfaces is hardly modelized by atomistic approaches, mostly because of increasing CPU time efforts for growing characteristic length scales of the simulated systems. In this situation, the phase field methods are relevant and, may be able to contribute as well [2]. Nevertheless, it was decided to use MD coarse-grained approach to modelize the separation behavior of polymer–polymer interfaces, through simulations of some connector chains embedded in two adjacent polymer entangled melts, see Fig. 1. Through the material force field between the particles, atomistic simulations are able to modelize internal cohesion forces in a material with a high level of accuracy indeed, more than Continuum Mechanics does. Atomistic (coarse-grained) approaches are thus helpfull and, contain exactly the molecular aspects needed to investigate and understand such a competition between viscoelastic deformation inside a polymer bulk-interphase and connector extraction out of this interphase.
Even though nice reviews on polymer adhesion can be found in the literature [3], [4] and some interesting theoretical and numerical predictions for mechanical behavior of separated interfaces and of strained polymer melts exist in the literature as well, see Refs. [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], the contribution of the viscoelastic deformation inside the polymers on either side of the interface are often disregarded or reduced to a simplified modeling. This last mechanism can be very important however, as the strain rate, the areal density or the connector chain length increase. This micro-bulk dissipation is not so obvious to determinate and, is taken under consideration in this manuscript.
In the manuscript proposed herein, our simulations of the failure of adhesion between two deformable melts develops a deeper insight in the competition between surface and bulk phenomena for a broader range of external conditions: this is the novelty of this contribution. Our investigations are aimed to improve further our study on the separate contributions of surface adhesion and bulk dissipation to the overall work of separation. This manuscript is divided into: (a), Section 2 describing some needful aspects on the computational efforts granted; (b), Section 3 synthesizing several aspects from the literature and proposing a short discussion on an over-simplified model derived from the literature; (c), Sections 4 Fracture kinematics of the polymer interfaces being modelized, 5 Some scaling laws of the polymer interfaces being modelized dealing with some kinematical aspects and scaling laws of the polymer interfaces being modelized. Finally, Section 6 presents the full interface separation behavior during pull-out and full debonding as an anticipation of the implementation of a new cohesive model, which will be the topic of further contributions to this manuscript.
Section snippets
Some aspects on the computational interface mechanics being modelized
Two amorphous polymer melts, stitched together by distinct connector molecules, are studied. This system is contained in an (initially orthorhombic) MD cell with a given cell matrix , see Fig. 1 and Ref. [16]. The melts are separated by an initial gap before any deformation, to impose the presence of an interface/ a craze being opened. The system is deformed by imposing the straining of the MD cell through a uniform rate of deformation gradient, see A. The principles of the
Legacy from the literature and short discussion
In the literature of adhesion (see Refs. [34], [35], [36], [37], [38]), the Universal Binding Energy Relation (UBER) describes the energy W dissipated (per unit of area) during the interface separation mechanism as a function of the separation (between the two surfaces).
Fracture kinematics of the polymer interfaces being modelized
In Section 2, we described how the cohesive behavior of our modelized interfaces was interpreted from MD simulation results. In Fig. 1, 2 one defined a (vertical) traction , representing an averaged measure of stress at the connector molecules, a
Some scaling laws of the polymer interfaces being modelized
Our present investigations are based on the velocity (or macroscopic strain rates) dependencies of (resp. W) as a function of the opening . They are presented in Fig. 14, Fig. 15, Fig. 16 for and , from where the results are combined into the results given in Fig. 13. An interesting first conclusion is that the “traction vs. opening” at the connectors during total debonding is still similar in shape to the “traction vs. opening” at the connectors during pull-out
Full interface separation behavior during pull-out and full debonding
One proposes here to analyze the cohesive law “ vs. ” during pure pull-out and full debonding, as a prelude to the mapping to a future continuum model for interface separation., although it will be the scope of a further contribution. During a pull-out simulation, it was already observed that as the strain rate increases, the two melts can be crushed against the intermediate walls, because of a faster extraction of the connectors. Thus, this artefact should make the extraction of the
Conclusion
The MD simulation approach was used to analyze the separation behavior of some connector chains embedded in two adjacent polymer entangled melts. We focused (a), on the pull-out mechanism, where the extraction of the connectors out of the melts takes place through a complex mechanism of forced reptation only and (b), on the full debonding process, where a micro-bulk dissipation contributes to the overall crack opening and can skew the description of the interface separation by pure connector
Data availability
The raw/processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study. Nevertheless, the modified routines used in LAMMPS will be given to the LAMMPS users-community after valuable publication.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgment
This contribution was the fruit of a long collaboration with Prof. Erik van der Giessen 7. This project was supported between 2010 and 2011 by the Dutch Technology Foundation STW under the project “A Multiscale approach toward integrated cohesive interface elements” and, between 2017 and 2020 by the Institut Charles Sadron (Team MIM, Strasbourg, FRANCE).
References (39)
- et al.
Int. J. Solids Struct.
(2009) - et al.
Polymer
(2020) - et al.
Comput. Mater. Sci.
(2020) - et al.
Comput. Theor. Polym. Sci.
(1998) Fast parallel algorithms for short-range molecular dynamics
J. Comput. Phys.
(1995)- et al.
J. Non-Newtonian Fluid Mech.
(2001) - et al.
J. Comput. Theor. Nanosci.
(2009) - et al.
J. Mech. Phys. Solids
(2002) - et al.
Int. J. Solids Struct.
(2003) van der Giessen
Comput. Mater. Sci.
(2012)
PRL
Physical Chemistry of Surfaces
Adhesion of Polymers
J. Chem. Phys.
Eur. Phys. Lett.
Macromolecules
Tamin Polym.
Cited by (4)
Protonation of Strained Epoxy Resin under Wet Conditions via First-Principles Calculations Using the H<sup>+</sup>-Shift Method
2023, Journal of Physical Chemistry BInterface damage and fracture mechanisms of a ceramic/polymer interface based on atomic-scale simulations
2022, Physical Chemistry Chemical PhysicsFirst-Principles Calculations of the Protonation and Weakening of Epoxy Resin under Wet Conditions
2021, Journal of Physical Chemistry BA discussion on a rate-dependent cohesive law for polymer adhesion inspired from existing constitutive equations
2021, World Congress in Computational Mechanics and ECCOMAS Congress