Dr. Bradley D. Olsen
Associate Professor
Dept. of Chemical Engineering,
MIT

Mellichamp Lecture

Biography
Bradley Olsen an Associate Professor in the Department of Chemical Engineering at MIT.  He earned his S.B. in Chemical Engineering at MIT, his Ph.D. in Chemical Engineering at the University of California – Berkeley, and was a postdoctoral scholar at the California Institute of Technology.  He started as an assistant professor at MIT in December 2009.  Olsen’s research expertise is in materials chemistry and polymer physics, with a particular emphasis on molecular self-assembly, block copolymers, polymer networks and gels, and protein biomaterials.  He has been recently recognized with the ACS Kavli Emerging Leader in Chemistry Lecture, the AIChE Colburn award, a Dreyfus Teacher-Scholar Award, and as one of CE&N Magazine’s Talented 12.

Abstract
Associative polymer networks are everywhere:  they are used to modify the texture of our food, the rheology of our consumer products, and to develop biomedical materials. They are used in enhanced oil recovery, and they also form the basis for many natural systems in our own bodies. While existing theories of these materials have been widely successful in explaining rheological behaviour, recent experiments from our group suggest that something fundamental is missing. In particular, measurements of diffusivity show a non-Fickian superdiffusive regime at length scales much larger than any structural scale in the gel. This behaviour is common across several chemically different polymer systems, suggesting it is a widespread phenomenon in associative polymers. The diffusion results can be empirically captured by a simple two-state reaction/diffusion model, but the connection to molecular properties is unclear. Therefore, we have developed a coarse-grained model for Brownian dynamics simulation that captures molecular diffusion, qualitatively reproducing the experimental behaviour and providing insight into the molecular origins of anomalous diffusion in associative polymers. Fundamentally, the new results originate from the fact that molecules may both hop and walk through the gel, with the difference being the number of sticky feet attached to the network. In the regime where skipping (a combination of hopping and walking) is dominant, superdiffusive scaling is observed. 

An intriguing application of these findings is in the development of new materials that can selectively filter and transport biomolecules. Nucleoporins are the family of proteins responsible for selective transport of biomolecules into and out of the nucleus, forming gel-filled pores in the nuclear membrane that control transport and protect the cell’s genetic cargo. Recently, we have shown that the transport function of the nuclear pore can be replicated with a minimal consensus repeat sequence derived from nucleoporins, yielding nucleoporin-like proteins (NLPs). Using concepts of associative polymer diffusion, we put forward a hypothesis to explain the function of these simplified NLPs based on the presence of walking and hopping diffusive modes in the gels. This hypothesis suggests design rules which allow the same transport properties observed in NLPs to be replicated in peptide-functionalized poly(ethylene glycol) (PEG) gels, completing the design cycle of adapting the performance of a natural material to a fully synthetic system based on insight into its fundamental transport mechanism. This shows great potential for advances in protein separation technology.