Topological Evolution of Polymer Networks

Plastics and polymers became one of the most widely used materials of the modern age. While in the previous decades, groundbreaking experimental discoveries fueled the spread of these materials and the growth of its industry, it still remained a difficult challenge to precisely describe the process of their formation.
In this project, computation methods are used to overcome challenges that arise in the experimental modelling of polymers. It is known that the structure (topology) of the networks formed by these materials can largely affect the physical properties of the final product. To have an accurate description of the polymer network's topology, it is required that in the bulk of the material, the connection pattern of the monomers between themselves are found.
While the set of tools available for experimental modelling are increasing by the day, it is still impossible to keep track of individual monomers during the polymerisation process experimentally. This on the other hand means no challenge for computational modelling. Utilising the fact that virtually any property can be found or measured during simulations - albeit with varying accuracy - Molecular Dynamics simulations are used to find the accurate description of short-range interactions between monomers as the polymerisation process proceeds.
In particular, the change in the reactivities of monomers are observed by using long simulations that model the curing process of different co-polymer melts. By having the capability of describing the driving forces of these simulation on a low level, it might be possible to find new ways of describing, and to virtually construct polymer networks. If this whole process is made effective enough, it could enable researchers to predict the behavior of polymers in extreme circumstances, or to find new, smartly designed materials.

Real polymer processes have a great deal of randomness, and as far as the topology is concerned, graphs can be used to give a precise description of them. These allow to translate a polymerisation process to an abstract Random Graph Simulations, where nodes represent the monomers and the edges between them the bonds. Nevertheless, a purely random simulation on a graph, where nodes to connect are selected with equal weights will result in unrealistic final structures.
This is due to the fact that in real systems, the surroundings of a monomer will affect its reactivity, which coresponds to the weight of nodes in the graph representation. With our methods, it is possible to find the accurate changes in the monomer reactivities, depending on its topological state. As a first approximation, this state depends solely on the degree of the monomer, that is how many other monomers it is connected to.
Having obtained the accurate evolution of monomer reactivities as a function of their topological state, it is possible to build polymer networks in a Molecular Dynamics environment, but with short-cutting the expensive explicit simulation of the curing process.

One particular application of the methods given above can be the modelling of the reverse process of polymerisation, depolymerisation, which in practice is the recycling of these materials. The ability to decompose plastics can also largely depend on the structure of the polymer's underlying network. By having realistic polymer networks in a virtual environment, it is also possible to computationally model, and explore new, more efficient ways in depolymerisation, which in turn can help reduce the footprint of the plastic industry significantly.