First-principle methods, such as density functional theory, are considered benchmark accurate, but are limited to small-size systems due to the demanding computational cost. Classic methods benefits from parameterizations are orders of magnitude faster in terms of computational efficiency, but lacks the precision in describing the microscopic structures that are often crucial to capture some subtle effects on the desired molecular properties. Our group applies multi-level models and develops new computational and theoretical tools to aid our studies. A few recently developed methods are presented in the following.
A hybrid atomistic electrodynamics/quantum mechanics approach: while classical electrodynamics simulations can accurately simulate the local electric field around metal nanoparticles, they offer few insights into the spectral changes that occur in SERS. First-principles simulations can directly predict the Raman spectrum but are limited to small metal clusters and therefore are often used for understanding the chemical mechanism. Therefore, we have developed a hybrid method bridging these two realms of computational approaches to facilitate the simulation of plasmon-enhanced spectroscopies, termed as DIM/QM. In short, a discrete interaction model (DIM) is used to represent the metal atoms, and the sample molecule remains treated by QM.
Theory of linear and non-linear vibrational spectroscopies: we have contributed to the development of theoretical approaches to understanding linear and nonlinear surface-enhanced vibrational spectroscopies. Specifically, we developed a unified description of enhancement mechanisms classified as either electromagnetic or chemical in nature is presented, termed as “dressed-tensors” formalism. The free molecular polarizabilities are “dressed” local field and field gradient tensors, which gives rise to the scattering polarizabilites. This method not only allows for fast computation, but also helps understand the spectral changes necessary for interpretation of linear and nonlinear surface-enhanced vibrational spectroscopies.
Subsystem density functional theory method: A combination of the calculation expense of large systems and the fact that most chemistry is centered around a smaller subsystem has led to the development of subsystem methods. Frozen density embedding (FDE) DFT allows for intuitive partitioning of the supermolecular system via the real-space electron density. This supermolecular system density is divided into the region of interest (“active” fragment) and the environment (“frozen” fragment).
Selected Publications: