Check out Dmitry’s new paper in JPCL, describing our new method for modeling nonadiabatic molecular dynamics in dense manifolds of electronic states.  In short, it does not require full knowledge of the many electronic PESs involved in the dynamics, but does not suffer from well-known problems arising from the mean-field nature of Ehrenfest dynamics.

Check out Wei-Tao’s newest paper in JPC C, describing an AIMD study of nanoparticle models containing multiple dangling bond defects.  We find that nuclear relaxation effectively uncouples neighboring defects, even at very short range, resulting in recombination dynamics that are nearly identical to those observed in isolated defects.

Better late than never:  Check out our recent perspective article in PCCP, arguing that the lowest energy conical intersections are inherently local feature of molecules.  That is, they arise from spatially local distortions of the geometric and electronic structure.  This locality has important consequences for photochemistry and photophysics.

Check out our group’s article in the 2019 Annual Review of Physical Chemistry covering our work on modeling conical intersections in semiconductor nanomaterials.

Check out Mike’s new paper in JPCA!  In it, we analyze the low propensity for nonradiative recombination in lead halide perovskites by identifying and characterizing conical intersections in two models of the surface of CsPbBr₃.  We only find conical intersections that are inaccessible upon excitation across the bulk band gap.  Analysis of the electronic structure around these intersections suggests that the ionic nature of the material contributes to the high energy of the intersections.

In Dmitry’s first paper in the group, he demonstrates that numerically exact solution of the vibrational Schrodinger equation in the adiabatic representation around a conical intersection in a finite basis is only possible if that basis contains discontinuous functions.  We demonstrate an approach based on a basis of discontinuous functions on a 2-D model and discussed how our approach could be extended to more dimensions.