Physics Colloquium, April 13, 2010
Traipsing through the ups and downs of rugged energy landscapes: chemical reactions and protein dynamics Rigoberto Hernandez

Georgia Institute of Technology

The magic of transition state theory (TST), as independently introduced by Marcelin, Wigner and Eyring, lies in the replacement of the dynamics of all the trajectories between reactants and products with a simple geometric calculation. Although the TST rate formula is not generally exact the order of magnitude of the rates is usually correct, and that has been of significant use in describing chemical reactions. In energy conserving systems, it is possible to do better. Rigorously non-recrossing dividing surfaces can be constructed from normal form theory, and have the structure of a normally hyperbolic invariant manifold. In dissipative systems, Kramers showed that naive transition state theory overestimates the rate at low friction (because reactants are not sufficiently activated) and at high friction (because reactants are slowed down before they reach the barrier). The mechanisms involved in these errors can be accounted for in modern transition state theory formulations, but the recrossing problem still persists. In recent work, we and others have constructed non-recrossing dividing surfaces appropriate for rate processes described by a Langevin equation. The catch is that the dividing surface is replaced by a family of time-dependent such surfaces with each specified to a given manifestation of the Langevin noise. The critical assumption that enabled this result is the fact that the Langevin noise does not depend on the reacting system. Can anything be done about solvents or baths that are more strongly coupled to the system? Yes. Moreover, these same ideas lend themselves to describing the dynamics of proteins between long-lived non-native structures.

Dr. Hernandez's Web Site

4:00 p.m., Physics Research Building (PRB), Room 1080

Reception at 3:45 p.m., Atrium, PRB