Novel Simulation Strategies For Predicting Polymer Properties

Project: Research project

Project Details



This award supports theory, computation, and education with an aim to advance understanding of polymeric materials. Polymeric materials are long chain-like molecules.

How sharp is that polymer interface? Important things happen at interfaces between different polymers. For example: hard plastic parts are stiff and tough to shatter, because they contain tiny droplets of rubbery polymer dispersed in a second, stiffer polymer. The rubbery droplets absorb impacts; without them, the plastic would crack when hit. This trick works because at the surface of the droplets, long polymer chains on each side of the interface intermingle enough to knit the interface together. Otherwise, the part could break easily, with the droplets pulling away from the stiff material as it cracked.

Another application for which polymer interfaces are crucial is in polymer-based solar cell materials. These materials are the subject of intense research focus as alternatives to 'hard' silicon-based photocells, with potential advantages including flexible materials and easier production; they are made like plastic sheeting in a chemical plant, not like computer chips in a billion-dollar ultraclean fab facility. Here, the interface is between special 'donor' and 'acceptor' semiconducting polymers. These donor-acceptor interfaces help pull apart pairs of opposite charges produced when sunlight is absorbed by the polymers, so the charges can be 'harvested' as electricity.

For rubber-toughened plastics, the interface needs to be a bit broad; for photocells, the interface needs to be sharp. For a given pair of polymers A and B, the interfacial width is controlled by how strongly the A and B chain repeat units repel each other. This repulsion depends on the chemical structure of A and B, which synthetic chemists can vary with amazing variety.

But how strongly will polymers of structures A and B repel each other? If this can be predicted, it could help the chemist decide what structures to make. The PI's research group has developed a new way to use computer simulations to predict the strength of repulsion between pairs of polymers, as a complement to challenging experiments. The PI's method relies on the idea that the more A and B repel each other, the more work is required to mix A and B together.

To compute the work of mixing, a trick inspired by CGI animation is used: start with a simulated sample of pure A chains, and gradually 'morph' half the chains into B chains, measuring how the energy rises as the morph progresses. By comparing the work required to prepare an A-B mixture by morphing half the chains, to the work to make a pure sample of B chains, by morphing all the chains, the work required to mix A and B chains can be measured.

The bigger the work of mixing is, the narrower the interface will be between demixed A and B material. In fact, the interface width is governed by a compromise between energy and entropy. Flexible polymer chains on either side of the interface have two conflicting desires: to avoid contact with the 'other' chains, and to wander about freely. The greater the cost for repulsive contacts, the more restricted are the wanderings of chains across the interface.

With the new 'morphing' simulation method, the PI can predict interfacial widths for polymer pairs from chemical structure alone. This powerful tool will help polymer chemists and engineers design more effective polymers for solar cells, as well a host of other applications in advanced materials.


This award supports theory, computation, and education with an aim to advance understanding of polymeric materials. With continuing improvements in computer power and recent advances in simulation techniques, it is finally becoming possible to predict key material properties for real polymers from their chemical structure. This project will develop and exploit new approaches to predict three kinds of properties:

1. interfacial tensions between polymer crystals, melts, and substrates, which are central to predicting nucleation barriers and designing nucleating agents;

2. chi parameters between polymer pairs, which are necessary for predicting miscibility, microphase separation, and strength of interfaces between immiscible polymers; and

3. entanglement length for flexible, semiflexible, and stiff polymer chains in melts and solutions, which is the central parameter in the modern theory predicting flow behavior of polymer melts and solutions from chain architecture.

Most polymers crystallize very reluctantly. Practical processing relies on nucleating agents and flow-induced crystallization to achieve high nucleation rates, resulting in materials with smaller spherulites and improved properties (optical clarity and strength). Predicting nucleation rates starts with classical nucleation theory, which describes the critical nucleus in terms of its free energy and tension with surrounding melt. Correspondingly, nucleating agents provide substrates for nuclei to form, with interfacial tensions that favor crystal over melt. To predict nucleation rates and design nucleating agents, it is critical to predict tensions between polymer crystals, melts, and substrates. In this project, new simulation methods will be developed to do that, for chemically realistic chains.

Predicting chi from chain structures is a grand unmet challenge for polymer theory. To account for how real molecules pack and interact, simulation is the right approach. Since chi measures the excess mixing free energy, free energies must be obtained from blend simulations. The PI has developed a new method, that computes the work to 'morph' one kind of chain into another. The PI will extend this method, from simple bead-spring chains to real chains in atomistic detail.

Multiple attempts have been made to construct theories for how the entanglement length scales with chain volume, flexibility, and concentration. Real polymer melts are well described by Lin-Noolandi scaling, stiff chains by Morse scaling. Everaers presented a scaling ansatz consistent with Morse but not LN. This project will develop a new scaling theory that unifies these approaches, and comprehensive simulations to measure entanglement across flexible and stiff chain melts and solutions.

The shale gas revolution is leading to substantial new investment in U.S. polyolefin plants. Better nucleating agents to improve properties of these low-cost ubiquitous polymers could lead to increased use as lightweight, energy-saving structural materials. Chi parameters between donor and acceptor chains in semiconducting polymers are hard to measure, but are key attributes in the design of heterojunction interfaces in next-generation polymer photovoltaics, which are the subject of intense research focus as potential complementary materials to hard silicon-based photocells, and may help enable the U.S. and the world to confront climate change and transition to a less carbon-intensive energy economy.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Effective start/end date9/1/198/31/22


  • National Science Foundation: $376,278.00


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.