Flow induced crystallisation in polymers: from molecules to processing

Polymer processing is a multi-billion pound, world-wide industry, manufacturing products used by virtually every person in the developed world (and beyond) on a daily basis. This vital sector of the UK economy will gain a significant competitive advantage from a molecular understanding of how polymers crystallise during processing, as it will enable stronger, lighter, more durable and more easily recycled plastic products. In this proposal we will overcome the key experimental, simulation and numerical issues in understanding polymer crystallisation to deliver a molecular based, predictive platform for the processing of semi-crystalline polymers. We will tightly integrate a family of progressively coarse-grained simulations and models, covering all relevant lengthscales within a single project. This will displace the current sub-optimal semi-empirical approaches in polymer processing and enable molecular design of polymer products, through choice of processing conditions. By facilitating the manufacture of polymer products with tailored properties this program will provide a critical competitive advantage to this important industry.

Polymers are long-chain molecules, formed from connecting together a large number of simple molecules. These long-chain molecules are at the heart of the multi-billion pound plastics industry. Semi-crystalline polymers make up a very significant fraction of the worlds production of synthetic polymers. Unlike simple molecules, the connectivity of polymer molecules means they crystallise into a composite structure of crystalline and amorphous regions. The proportion of amorphous and crystalline material, along with the arrangement and orientation of the crystals, is collectively known as the morphology. The crystal morphology strongly influences strength, toughness, permeability, surface texture, transparency, capacity to be recycled and almost any other property of practical interest. Furthermore, polymer crystallisation is radically influenced by the flows that are ubiquitous in polymer processing. Flow drastically enhances the rate at which polymers crystallise and has a profound effect on their morphology. Flow distorts the configuration of polymer chains and this distortion breaks down the kinetic barriers to crystallisation and directs the resulting morphology.

Understanding polymer crystallisation is a formidable problem. The huge range of relevant lengthscales ranges from the size of a monomer (nm) up to near macroscopic crystals (micro-metres). The range of timescales is even wider, ranging from the monomer relaxation time (ns) to nucleation (hours at low under-cooling). Our project will involve extensive multiscale modelling, supported at each level by experiments specifically designed to address key modelling issues. Our experiments will involve controlled flow geometries, the systematic variation of molecular weight and the probes of both nucleation and overall crystallisation. Close integration of experiments and all levels of modelling is a key feature.

We will develop an interrelated hierarchical family of multiscale models, spanning all relevant lengthscales and delivering results where piecewise approaches have been ineffective. Each technique will be tightly integrated with its neighbours, retaining the molecular basis of the models while progressively addressing increasingly challenging systems. This will cumulate with the low-undercooling and high-molecular weights that are characteristic of polymer processing. Each simulation will use a rare event algorithm to dramatically increase the nucleation rate, the cause of the very long timescales. Insight from the most detailed models will guide the development of faster modelling. At the highest coarse-graining, the program will derive models suitable for computational modelling of polymer processing. Using these models in cutting-edge finite element code, we will compute FIC behaviour in polymer processing geometries.

Flow-induced crystallization in polymers is a fascinating example of an externally driven, nonequilibrium phase transition, controlled by kinetics. Furthermore, FIC is ubiquitous in industrial processing of semicrystalline polymers, the largest group of commercially useful polymers. Hence the research described herein will be impact significantly on both fundamental and industrial polymer science and engineering. Furthermore, the experimental, simulation and modelling techniques developed in the project will impact on the many fields where molecular motion determines macroscale properties of practical importance.

Academic
The most important academic impacts from this project will arise from the novel results from experiments, simulation and modelling of polymer FIC. Furthermore the tools for rare-event simulation, coarse-graining and multiscale modelling developed herein will impact strongly across the numerous fields that depend on molecular modelling. Full details of academic impacts are given in the Academic Beneficiaries section.

Industrial and technological
Industrial polymer processors will benefit significantly from the control and customisation of polymer processing and products that our modelling will deliver. In particular:
- Practical computational tools from the project will accurately predict crystal morphology from processing conditions, this will enable process engineers to design and optimise the properties of polymer products by tailoring processing speed, temperature, molecular weight and geometry.
- The ability to control crystal morphology will allow process engineers to manufacture products that are lighter, stronger, more transparent, less permeable and easier to recycle using processes that are faster, more efficient and less prone to instabilities, all of which will increase profitability.
- The ability to optimise solid-state properties will make polymer products suitable for a greater range of product lines, opening new markets and revenue streams to the polymer industry.
- Software developers for polymer processing will benefit from the predictive ability of our models, combined with our techniques in computational fluid dynamics. The resulting improved speed, accuracy and predictive ability will lead to wider uptake and exploitation of their polymer process modelling software.

Societal and general public
Polymer processing is a multi-billion pound, world-wide industry, manufacturing products used by virtually every person in the developed world (and beyond) on a daily basis. This vital sector of the UK economy will gain a significant competitive advantage from a molecular understanding of how polymers crystallise during processing, as it will enable stronger, lighter, more durable and more easily recycled plastic products. There will be environmental benefit from more efficient packaging and greater recycling.

Wider academic impacts
Beyond the project's central aim there are numerous impacts from the cutting-edge molecular simulation and modelling techniques developed herein. Molecular simulation and modelling techniques can predict a wide-range of physical properties, relevant to numerous academic and technological fields. Examples include viscosity, heat capacity, latent heat, solubility and diffusion rates. Our novel simulation tools provide a flexible framework for highly efficient molecular simulation of phase transitions and barrier crossing. This will facilitate rapid characterisation of many molecular systems, with the ability to efficiently explore a wide range of temperature being an especially attractive feature. Examples of fields this will impact on include polymers, colloids, protein folding, drug delivery and molecular self-assembly.