Funded Grants

Emergence of complex material structures from particle assembly

A child presented with a bin full of wooden building blocks can entertain itself for hours as it delights in creating caricatures of things it has seen. Even with only a few different shapes, familiar objects soon materialize as the basic blocks are combined to form castles, cars, horses and houses. A wicked babysitter can quickly turn the child’s delight to frustration by hiding magnets in a few of the blocks. All of a sudden, the rules that worked to build a tower no longer hold true! Some blocks unexpectedly stick together, pulling key supports out of place and toppling whole towers! Other blocks repel, making it impossible to orient them in the desired fashion. While this diabolical trick on a child seems unnecessarily cruel, it is the sort of challenge faced by engineers working to self-assemble structures on the molecular level.

Particles on length scales from one millionth to one billionth of a meter can have very strong and complex interactions with each other- so complex that they make working with magnetized blocks seem like child’s play. These so-called nanoparticles have an infinitely tailorable set of interactions made possible by advances in synthetic chemistry. Tethers of long molecules called polymers can be attached to nanocubes, triangles, ellipsoids, and polyhedra. The surfaces of these particles can be modified with sticky or repulsive patches and we are on the cusp of being able to create whole new classes of particles with switchable interactions. Adding to this library is an enormous set of particles provided to us by Nature in the form of proteins and other biological macromolecules. Nature has also shown us that these building blocks can be combined to create amazingly useful structures including viruses, mitochondria, and bones. The goal of our research is to help build an understanding of how Nature accomplishes this feat of engineering. Our ability to fabricate materials and devices from the bottom up hinges upon understanding how the forces between building blocks conspire to produce emergent patterns and how to exploit these forces to make tiny particles self-assemble into the patterns we desire.

Currently there exists no theory that predicts the structures assembled by a system of complex interacting particles. Furthermore, there currently exists no way to determine what structural properties might emerge when these particles organize. Developing this theory of self-assembly is crucial to engineering the next generation of materials and devices in which nanoparticles are used in place of atoms or molecules as building blocks. An example of a question we may ask as we build this theory is what are the necessary interactions between 60 identical particles that will cause them to robustly form an icosahedron, the shape of many common viruses? Answering it will allow us to develop synthetic virus capsids that can be used as drug delivery devices, as well as provide insights into how we may prevent malevolent viruses from assembling in our own bodies. The applications for bottom-up materials design are as diverse as the creatures found in Nature. Imagine fabrics that automatically control the temperature and humidity of the skin, self-healing fiber-optic cables with stronger signals, and solar cells that grow on the roofs of buildings. A theory of self-assembly would help bring us closer to making these dream devices a reality.

The current computational and experimental methods used to study self-assembly are not sufficient for building the necessary theory. Direct experimentation on systems of such particles can be expensive and yield only one structure per trial. Searching for new structures and morphologies via this trial-and-error approach is both cost and time prohibitive. On the other side of the science wing, the thermodynamic simulations used to predict crystalline structures in various equilibrium systems are also at a loss to deal with the nonlinear, conditional interactions that dominate complex systems. It is clear that new methods are necessary in order to usher in the next generation of smart materials.

We propose creating new computer models of these systems to create a bridge between theory and experiment in order to develop our understanding of how complicated nanoparticles selfassemble. Modeling such particles in silico enables us to test hypotheses about how we believe particles behave much more quickly than can be achieved in a laboratory. This agility will help us focus on the characteristics that play an important role in the assembly of desired structures, which will in turn allow us to guide experiments in the lab. Borrowing from biology again, we can mimic the evolution of self-assembling proteins in the computer to help us design particles that will generate target structures. That is, if the degree to which a system of particles assembles a desired structure can be quantified then a genetic algorithm can be used to find the traits of particles that do an excellent job at forming the structure in question.

Through building a bridge between theory and experiment we will not only set the stage for engineering next generation materials and nanodevices, but we will also enhance our understanding of complex biological self-assembly. My group at the University of Michigan is uniquely poised to make advances in this nontraditional approach to soft-matter simulation due to our expertise in self-assembly and our association with UM’s Center for the Study of Complex Systems. The first realistic model of switchable nanoparticles suspended in a solution is currently under development in my group by the same student who is pioneering the use of genetic algorithms to find potentials that best generate crystal lattices. Furthermore, our experience in defining metrics such as customized order parameters involving spherical harmonics will be invaluable as we develop fitness functions based on structure factors for use in the genetic algorithm. During this course of research, development of new algorithms and tools for studying self-assembling systems will be at the forefront of our attention, with the intention being that these tools might find extended applications in related complex systems in other disciplines.